Personalized recommendation method, system and equipment for medical service and storage medium

Abstract

The invention discloses a personalized recommendation method, system and device for medical services and a storage medium, and relates to the field of medical information. In the method, medical insurance use data of a user is acquired, and the medical insurance use data comprises a historical medical record, an expense settlement record, a medicine purchase record and a physical examination report; extracting target features associated with medical insurance use habits and demands from the medical insurance use data, wherein the target features comprise medical treatment frequency, medical treatment types, expense types, disease types and medicine use habits; using a preset deep learning model to predict future medical seeking demands and cost budget of the user according to the target features; and recommending a target hospital and a target doctor according to the future medical seeking demand and the cost budget. By implementing the technical scheme provided by the invention, the medical experience of the patient is optimized.

Description

Technical Field

[0001] The present application relates to the technical field of medical information, and in particular to a method, system, device and storage medium for personalized recommendation of medical services. Background Art

[0002] As a bridge between medical institutions and patients, the medical insurance service platform not only simplifies the medical service process, but also greatly improves the user experience. Through these platforms, users can easily obtain personal medical records, fee settlement details, drug purchases, physical examination reports and other information, greatly improving the transparency and convenience of medical services.

[0003] In the existing medical insurance services, in order to improve the user experience, major platforms have adopted a variety of means to achieve personalized services. The common practice is to recommend nearby hospitals and doctors based on the user's location, or directly recommend hospitals and doctors that the user has visited in the past. These practices cannot effectively capture the user's personalized characteristics, resulting in inaccurate recommendation results and poor user experience.

[0004] Therefore, how to more effectively capture the personalized characteristics of users and improve the recommendation accuracy of the medical insurance service platform has become a technical problem that needs to be solved urgently. Summary of the invention

[0005] The present application provides a personalized recommendation method, system, device and storage medium for medical services, which can recommend hospitals and doctors that better meet the needs of users, thereby improving the accuracy and efficiency of medical services.

[0006] In a first aspect of the present application, a personalized recommendation method for medical services is provided, which is applied to a medical insurance service platform, and the method comprises: Obtain the user's medical insurance usage data, which includes historical medical records, fee settlement records, drug purchase records, and physical examination reports; Extracting target features associated with medical insurance usage habits and needs from the medical insurance usage data, the target features including medical frequency, medical type, expense type, disease type, and medication habits; Using a preset deep learning model to predict the user's future medical needs and cost budget according to the target features; Based on the future medical needs and cost budget, recommend target hospitals and target doctors.

[0007] Optionally, the target hospital and target doctor recommended according to the future medical needs and cost budget include: According to the future medical needs and cost budget, all hospitals in the current area are screened to obtain a first hospital set; Score all hospitals in the first hospital cluster according to their geographical location, professional characteristics, doctor resources, and equipment and facilities to obtain a basic score; Determine a historical medical hospital according to the historical medical records, and score the historical medical hospitals concentrated in the first hospital according to the medical insurance usage data and the user's evaluation to obtain additional points; Obtaining a total hospital score according to the basic score and the additional score, and screening the first hospital set according to the total hospital score to obtain a second hospital set; All doctors in the second hospital that meet the medical field requirements are scored based on doctor grade, scheduling and patient evaluation index to determine the target doctor.

[0008] Optionally, the scoring of all hospitals in the first hospital set according to the hospital's geographical location, professional characteristics, doctor resources, and equipment and facilities to obtain a basic score includes: The first score is determined based on the geographical location of the hospital and the commuting time and commuting cost of the target location preset by the user, and the second score is determined based on the matching degree between the hospital's specialty departments and the user's future medical needs; Determine a target department according to the future medical needs, determine a third score according to the richness of the doctor resources of the target department of the hospital, and determine a fourth score according to the perfection of the equipment and facilities associated with the target department in the hospital; The first score, the second score, the third score, and the fourth score are added together to obtain the basic score.

[0009] Optionally, scoring the historical medical hospitals in the first hospital cluster according to the medical insurance usage data and the user's evaluation index to obtain additional points includes: Utilize natural language processing technology or machine learning algorithms to analyze the diagnostic information in the historical medical records, identify the type of disease the user suffers from and the severity of the disease, and determine the trend of disease changes based on physiological data at different time points in the historical medical records; determining a fifth score according to the disease type, the severity and the disease change trend, and determining a sixth score according to the user's evaluation index; The fifth score and the sixth score are added together to obtain the additional score.

[0010] Optionally, the step of scoring all doctors in the second hospital that meet the medical field based on doctor grades, shift schedules, and patient evaluation indexes to determine target doctors includes: Determine a target medical field according to the future medical needs, and screen the doctors in the second hospital set according to the target medical field to obtain a set of doctors to be selected; The seventh score is determined based on the doctor's professional title and years of practice, the eighth score is determined based on the doctor's appointment availability and average waiting time, and the ninth score is determined based on the patient evaluation index; The seventh score, the eighth score and the ninth score are weightedly summed to obtain a doctor score, and the doctor with the highest doctor score is determined as the target doctor.

[0011] Optionally, the method further includes: Determine the user's medication habits based on the medical insurance usage data, where the medication habits include drug purchase records, medication frequency, and medication cycle; Building a health profile of the user based on the medication habits and physical examination report; In combination with medical knowledge and health guidelines, health guidance suggestions are provided to the user based on the health portrait.

[0012] Optionally, the health guidance suggestions provided to the user based on the health portrait in combination with medical knowledge and health guidelines include: Assess the disease risk of the user according to the health profile, and determine the target disease whose disease risk is higher than a threshold; Personalized dietary advice is provided based on the influencing factors of the target disease and the physiological indicators of the user, wherein the dietary advice includes recommended food types, intake amounts, and cooking methods.

[0013] In a second aspect of the present application, a personalized recommendation system for medical services is provided, including a collection module, a feature module, a prediction module and a recommendation module, wherein: A collection module configured to obtain the user's medical insurance usage data, wherein the medical insurance usage data includes historical medical records, fee settlement records, drug purchase records, and physical examination reports; a feature module configured to extract target features associated with medical insurance usage habits and needs from the medical insurance usage data, the target features including medical frequency, medical type, expense type, disease type, and medication habits; A prediction module, configured to use a preset deep learning model to predict the user's future medical needs and cost budget according to the target features; The recommendation module is configured to recommend a target hospital and a target doctor based on the future medical needs and cost budget.

[0014] In the third aspect of the present application, an electronic device is provided, including a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes any one of the methods described above.

[0015] In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores instructions, and when the instructions are executed, any of the methods described above is executed.

[0016] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. By acquiring and analyzing the user's medical insurance usage data, including historical medical records, expense settlement records, drug purchase records and physical examination reports, we can have a more comprehensive understanding of the user's medical needs and habits; extracting information related to the target characteristics, such as medical frequency, medical type, expense type, disease type and medication habits, helps to more accurately predict the user's future medical needs and expense budget; based on these prediction results, we can recommend hospitals and doctors that better meet the user's needs, thereby improving the accuracy and efficiency of medical services; 2. Personalized recommendation methods can guide users to choose hospitals and doctors more reasonably, avoiding blind medical treatment and waste of medical resources. By predicting users' future medical needs, medical institutions can make resource allocation and preparations in advance to cope with possible medical peaks, thereby optimizing the allocation of medical resources. 3. Personalized recommendation methods can provide users with medical services that better meet their needs, thereby improving their medical experience and satisfaction. Users can choose more suitable hospitals and doctors based on their actual conditions and preferences, reducing the inconvenience and troubles in the medical process; 4. By introducing advanced technologies such as deep learning, the medical insurance service platform can better understand and serve users and provide more intelligent and personalized medical services; by analyzing users' medical insurance usage data, medical insurance management departments can formulate and adjust medical insurance policies more scientifically and improve the efficiency of medical insurance funds; at the same time, personalized recommendation methods can also help medical insurance management departments better understand users' medical needs and habits, and provide more powerful data support for the formulation of medical insurance policies. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a flowchart of a personalized recommendation method for medical services disclosed in an embodiment of the present application; Figure 2 It is a module diagram of a personalized recommendation system for medical services disclosed in an embodiment of the present application; Figure 3 It is a structural schematic diagram of an electronic device disclosed in an embodiment of the present application.

[0018] Explanation of the reference numerals: 201, acquisition module; 202, feature module; 203, prediction module; 204, recommendation module; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. DETAILED DESCRIPTION

[0019] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0020] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.

[0021] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0022] This embodiment discloses a personalized recommendation method for medical services, which is applied to a medical insurance service platform. Figure 1 is a flow chart of a personalized recommendation method for medical services disclosed in an embodiment of the present application, such as Figure 1 As shown, the method comprises the following steps: S110, obtaining the user's medical insurance usage data, wherein the medical insurance usage data includes historical medical records, fee settlement records, drug purchase records, and physical examination reports; Medical insurance usage data belongs to personal privacy data, and the medical insurance usage data obtained in the embodiments of this application are authorized by the user. Historical medical records refer to detailed records of users going to medical institutions (such as hospitals, clinics, etc.) for medical services due to health problems in the past period of time. These records usually include the date of consultation, the department of consultation, the diagnosis result, the doctor's order, the treatment process, etc. Historical medical records are very important for doctors to understand the patient's medical history, health status and treatment effect, and help to formulate a more reasonable diagnosis and treatment plan. Cost settlement records refer to detailed records of users settling costs with medical insurance institutions or medical institutions after receiving medical services. These records usually include medical expenses, medical insurance payment amounts, personal payment amounts, payment methods, etc. Cost settlement records help users understand their medical expenses, and are also an important basis for medical insurance institutions to review and settle expenses. Drug purchase records refer to detailed records of users purchasing drugs through medical insurance designated pharmacies or other channels. These records usually include purchase date, drug name, specifications, quantity, price, purchase method, etc. Drug purchase records help users understand their medication, and are also one of the bases for medical insurance institutions to review and settle drug expenses. Physical examination reports refer to detailed health reports obtained by users through regular physical examinations. These reports usually include measurement results of various physical examination indicators (such as blood pressure, blood sugar, blood lipids, etc.), abnormal prompts, doctor's advice, etc. Physical examination reports help users understand their health status, discover potential health problems in time, and take corresponding preventive measures. At the same time, physical examination reports are also an important reference for doctors to formulate diagnosis and treatment plans.

[0023] S120, extracting target features associated with medical insurance usage habits and needs from the medical insurance usage data, the target features including medical frequency, medical type, expense type, disease type, and medication habits; The frequency of medical treatment refers to the number of times a user visits a medical institution within a certain period of time (such as one year, half a year or quarter). By counting the user's historical medical records and calculating the number of medical treatments within a certain period of time, the frequency of medical treatment can be obtained. The frequency of medical treatment reflects the user's health status and medical treatment habits. Frequent medical treatment may mean that the user has a chronic disease or health problem, while less medical treatment may indicate that the user is in good health or has a weak awareness of medical treatment. The type of medical treatment refers to the type of medical institution (such as general hospital, specialist hospital, community hospital, etc.) and department (such as internal medicine, surgery, pediatrics, etc.) selected by the user when visiting the doctor. According to the user's historical medical records, the type of medical institution and department information selected by the user each time he visits the doctor are extracted. The type of medical treatment reflects the user's medical treatment preference and disease type. For example, users who choose specialist hospitals may suffer from a specific disease, while users who choose community hospitals may prefer routine physical examinations or diagnosis and treatment of minor diseases. The type of expense refers to the type of expense incurred by the user when receiving medical services, including medical insurance payment, personal payment, commercial insurance payment, etc. By analyzing the user's expense settlement records, the expense type and amount of each visit are extracted. The expense type reflects the user's ability to pay and the impact of medical insurance policies on it. Understanding the expense type helps to formulate more reasonable expense payment plans and medical insurance policy adjustments. Disease type refers to the type and severity of the user's disease. Based on the user's historical medical records and diagnosis results, the name and classification of the disease they suffer from are extracted. Disease type reflects the user's health status and medical needs. Understanding disease type helps to formulate more personalized treatment plans and health guidance suggestions. Medication habits refer to the user's preferences and regularities when using drugs, including drug types, frequency of use, dosage, etc. By analyzing the user's drug purchase records and medical advice information, their medication habits and preferences are extracted. Medication habits reflect the user's dependence on drugs and the standardization of medication. Understanding medication habits helps to formulate more reasonable medication plans and avoid drug abuse or misuse.

[0024] S130, using a preset deep learning model to predict the user's future medical needs and cost budget according to the target features; Select an appropriate deep learning model based on the type of target features and the requirements of the prediction task. Commonly used models include convolutional neural networks (CNN), recurrent neural networks (RNN), long short-term memory networks (LSTM), autoencoders, and deep neural networks (DNN). For medical demand prediction, you may need to consider information such as the user's historical medical records and disease type, and choose a model that can process time series data or has strong feature extraction capabilities. For cost budget prediction, you may need to consider factors such as the user's cost type, medical type, and disease severity, and choose a model that can process multivariate data and output continuous values.

[0025] Perform preprocessing operations such as cleaning, normalization, and standardization on the target features to improve the training efficiency and prediction accuracy of the model. Construct training sets, validation sets, and test sets for model training, validation, and testing. Use the training set data to train the deep learning model and update the model parameters through the back propagation algorithm. During the training process, cross-validation, early stopping, and other techniques can be used to prevent the model from overfitting. According to the performance on the validation set, adjust hyperparameters such as learning rate, batch size, and network structure to optimize model performance.

[0026] Use target features (such as frequency of medical treatment, type of medical treatment, type of disease, etc.) as input to the deep learning model. For time series data, such as medical records, the model can be input in chronological order. The deep learning model automatically learns the rules and patterns in the data based on the input features. Use the trained model to predict medical needs in the future, such as predicting the type of disease a user may suffer in the next few months, the number of medical treatments, etc. Explain and interpret the model's prediction results to provide valuable medical decision support. Visualization techniques can be used to display the prediction results, such as drawing prediction curves, bar charts, etc.

[0027] Take features related to the target cost (such as cost type, type of medical treatment, severity of disease, etc.) as input to the deep learning model. Consider the user's historical cost data to more accurately predict future cost budgets. The deep learning model learns the rules and patterns in the cost data based on the input features. Use the trained model to predict medical expenses in the future, such as predicting the total amount of medical expenses that users may incur in the next few months or the distribution of various types of expenses. Analyze and evaluate the model's prediction results to understand the accuracy and reliability of the cost budget. More reasonable cost payment plans and medical insurance policy adjustments can be formulated based on the prediction results.

[0028] S140. Recommend target hospitals and target doctors based on the future medical needs and cost budget.

[0029] Based on the prediction results of the deep learning model, we can understand the types of diseases that users may suffer in the future, the frequency and type of medical treatment, and other needs. These needs will serve as an important basis for recommending target hospitals and target doctors. Based on the user's expense type and historical expense data, we can predict the cost budget that users may incur in the future medical treatment process. The cost budget will help screen hospitals and doctors that meet the user's economic affordability.

[0030] According to the user's medical needs, such as disease type, medical type (such as emergency, outpatient, inpatient, etc.), select qualified hospitals. Consider the hospital's geographical location, medical level, professional characteristics, patient evaluation and other factors to further narrow the selection range. Combined with the user's cost budget, select hospitals that can meet the user's needs in terms of price and service. Among the selected hospitals, find doctors with relevant professional knowledge and experience according to the user's disease type and medical needs. Consider the doctor's professional background, education, title, patient satisfaction, academic achievements and other factors, and select excellent doctors for recommendation. It can further provide users with more convenient medical advice based on factors such as the doctor's scheduling and appointment difficulty. Display the selected hospital and doctor information to users in the form of a list or map, etc., so that users can view and choose. Provide detailed introductions to hospitals and doctors, including professional characteristics, service scope, contact information, etc., to help users make decisions.

[0031] Collect user feedback on the recommendation results to understand user satisfaction with hospitals and doctors. Adjust recommendation algorithms and strategies based on user feedback to improve the accuracy and satisfaction of recommendations. Regularly update information on hospitals and doctors, including newly added hospitals and doctors, changes in doctors' professional directions, etc. Dynamically adjust recommendation results based on changes in users' medical needs and cost budgets. Provide users with more personalized recommendation services based on user personal preferences, medical history, and other information. For example, recommend doctors that better meet users' needs based on users' past medical records and doctor reviews. Collect user satisfaction with recommendation services through questionnaires, online reviews, and other methods. Analyze users' acceptance and satisfaction with recommendation results to understand the actual effect of recommendation services. Analyze users' medical behavior and cost expenditures to evaluate the impact of recommendation services on users' medical decisions and cost management. Based on the analysis results, continuously optimize recommendation algorithms and strategies to improve the efficiency and effectiveness of recommendation services.

[0032] Optionally, the target hospital and target doctor recommended according to the future medical needs and cost budget include: According to the future medical needs and cost budget, all hospitals in the current area are screened to obtain a first hospital set; Score all hospitals in the first hospital cluster according to their geographical location, professional characteristics, doctor resources, and equipment and facilities to obtain a basic score; Determine a historical medical hospital according to the historical medical records, and score the historical medical hospitals concentrated in the first hospital according to the medical insurance usage data and the user's evaluation to obtain additional points; Obtaining a total hospital score according to the basic score and the additional score, and screening the first hospital set according to the total hospital score to obtain a second hospital set; All doctors in the second hospital that meet the medical field requirements are scored based on doctor grade, scheduling and patient evaluation index to determine the target doctor.

[0033] Based on the prediction results of the deep learning model, the user's medical needs in the future are clarified, including the type of disease, type of medical treatment (such as outpatient, hospitalization, surgery, etc.) and the expected frequency of medical treatment. At the same time, based on the user's expense type and historical expense data, the cost budget that the user may incur in the future medical treatment process is predicted. According to the user's geographical location and medical needs, all hospitals in the current area are preliminarily screened. The screening conditions may include the hospital's geographical location (such as distance from the user), professional characteristics (such as whether there are relevant departments or professional teams), doctor resources and equipment facilities, etc. The screening results form the first hospital set, which includes hospitals that meet the basic needs of users.

[0034] Establish a comprehensive scoring system to score all hospitals in the First Hospital. Scoring indicators may include the hospital's geographical location (such as transportation convenience), professional characteristics (such as department strength, scientific research results), doctor resources (such as the number of doctors, professional titles, academic qualifications, etc.), equipment and facilities (such as medical equipment, ward conditions, etc.). According to the above scoring system, all hospitals in the First Hospital are scored to obtain the basic score of each hospital. The basic score reflects the comprehensive performance of the hospital.

[0035] Analyze the user's historical medical records to determine the hospitals that the user has visited in the past (historical medical hospitals). For the historical medical hospitals that are concentrated in the First Hospital, additional scores are given based on the medical insurance usage data and user evaluations to obtain additional points. The additional points may include factors such as medical insurance reimbursement ratio, user satisfaction, and historical medical visits. The additional points reflect the user's preference and satisfaction with the historical medical hospitals.

[0036] The basic score and additional score of each hospital are added together to obtain the total score of the hospital. The total score of the hospital comprehensively reflects the performance of the hospital in comprehensive aspects and user preferences. The first hospital set is sorted and screened according to the total score of the hospital to obtain the second hospital set. The second hospital set contains hospitals that perform well in both comprehensive aspects and user preferences.

[0037] Establish a doctor rating system to rate all doctors in the second hospital who meet the medical field. The rating indicators may include doctor level (such as professional title, education), scheduling (such as available appointment time, scheduling density), patient evaluation index (such as satisfaction, treatment effect, etc.). According to the above rating system, all doctors in the second hospital who meet the medical field are scored, and the target doctor is determined based on the scoring results. The target doctor is a doctor who performs well in the professional field, is convenient to schedule, and has a high patient rating. The selected second hospital set and target doctor information are displayed to users in the form of a list or map, etc., for easy viewing and selection. Provide a detailed introduction of the hospital and doctors, including professional characteristics, service scope, contact information, etc., to help users make decisions.

[0038] The embodiment of the present application not only takes into account the user's future medical needs and cost budget, but also combines the hospital's geographical location, professional characteristics, doctor resources, equipment and facilities and other dimensions for scoring, so as to more comprehensively evaluate the comprehensive strength of the hospital. At the same time, by additionally scoring the historical medical hospital, the user's historical medical experience and evaluation can be fully utilized to further improve the accuracy of the recommendation. The embodiment of the present application can provide users with more personalized recommendation services based on the user's personal preferences, medical history and other information. This helps to meet the personalized needs of users and improve user satisfaction and medical experience. Through accurate recommendations, patients can be guided to be reasonably diverted to avoid excessive concentration on certain popular hospitals or doctors, thereby improving the utilization efficiency of medical resources. This helps to alleviate the problem of tight medical resources and improve the overall level of medical services. The embodiment of the present application can motivate hospitals and doctors to continuously improve their medical service quality and professional level through scoring and recommending hospitals and doctors. This helps to form a benign competition mechanism and promote the healthy development of the medical industry. The embodiment of the present application can provide users with more convenient medical services through an intelligent recommendation system. Users only need to enter their medical needs and cost budget to get recommendations for hospitals and doctors that meet their needs, which greatly saves users' time and energy. By analyzing users' medical needs and cost budgets, data support and reference can be provided for the formulation of medical insurance policies. This helps medical insurance policies meet users' needs more accurately and improves the efficiency of medical insurance funds.

[0039] Optionally, the scoring of all hospitals in the first hospital set according to the hospital's geographical location, professional characteristics, doctor resources, and equipment and facilities to obtain a basic score includes: The first score is determined based on the geographical location of the hospital and the commuting time and commuting cost of the target location preset by the user, and the second score is determined based on the matching degree between the hospital's specialty departments and the user's future medical needs; Determine a target department according to the future medical needs, determine a third score according to the richness of the doctor resources of the target department of the hospital, and determine a fourth score according to the perfection of the equipment and facilities associated with the target department in the hospital; The first score, the second score, the third score, and the fourth score are added together to obtain the basic score.

[0040] Get the commuting time between each hospital and the user's preset target location (such as home address, workplace, etc.). This can be achieved through the map service API (application programming interface) or traffic database. The commuting time can be set as a scoring standard. For example, the shorter the commuting time, the higher the score; conversely, the longer the commuting time, the lower the score. In addition to the commuting time, you can also consider the commuting cost, such as public transportation costs, self-driving fuel costs, etc. Similarly, the commuting cost can also be set as a scoring standard. The lower the cost, the higher the score; the higher the cost, the lower the score. The commuting time and commuting cost are comprehensively evaluated according to the set weights to obtain the first score of each hospital in terms of geographical location. Understand the departments that each hospital is good at through the hospital's official website, medical database or professional medical evaluation agency. Match the user's future medical needs (such as disease type, medical type, etc.) with the hospital's departments that are good at. The higher the match, the higher the score; the lower the match, the lower the score. Determine the target department based on the user's future medical needs. Statistics are collected on the number of doctors, doctor titles (such as chief physician, deputy chief physician, etc.), and doctor education in the target department of each hospital. Based on the statistical information of doctor resources, the richness of doctor resources in the target department of each hospital is evaluated. Hospitals with more doctors, higher titles, and higher education have higher scores. Obtain a list of equipment and facilities associated with the target department of each hospital, including information such as equipment name, model, and quantity. Based on the list of equipment and facilities, evaluate the completeness of equipment and facilities in the target department of each hospital. The more advanced and complete the equipment and facilities, the higher the score; conversely, the backward or insufficient the equipment and facilities, the lower the score. Summarize the scores of each hospital in terms of geographical location, professional characteristics matching, richness of doctor resources, and completeness of equipment and facilities. According to the actual situation, different weights can be set for the scores of different aspects to reflect their importance when users choose hospitals. According to the set weights, the scores of each aspect are weighted and summed to obtain the basic score of each hospital.

[0041] For example, user Mr. Zhang lives in the city center and his future medical needs are for heart problems. The First Hospital includes three hospitals: A, B, and C.

[0042] Hospital A: commuting time is 15 minutes, and commuting cost is 5 yuan. According to the scoring criteria, commuting time within 15 minutes will get 5 points, and cost within 10 yuan will get 5 points, so Hospital A gets 10 points in terms of geographical location. Hospital B: commuting time is 30 minutes, and commuting cost is 10 yuan. According to the scoring criteria, commuting time between 15-30 minutes will get 3 points, and cost within 10 yuan will get 5 points, so Hospital B gets 8 points in terms of geographical location. Hospital C: commuting time is 45 minutes, and commuting cost is 20 yuan. According to the scoring criteria, commuting time exceeding 30 minutes will get 1 point, and cost exceeding 10 yuan will get 1 point, so Hospital C gets 2 points in terms of geographical location. Hospital A: Its expertise includes cardiology, which fully matches Mr. Zhang's future medical needs, and it gets 10 points. Hospital B: Its expertise includes neurology and orthopedics, which do not match Mr. Zhang's cardiology needs, and it gets 0 points. Hospital C: Its expertise includes cardiology and respiratory medicine, which match Mr. Zhang's cardiology needs, and it scored 10 points. Hospital A: It has abundant cardiology physician resources, including several chief physicians and deputy chief physicians, and it scored 10 points. Hospital B: It has scarce cardiology physician resources, with only one attending physician, and it scored 2 points. Hospital C: It has relatively abundant cardiology physician resources, including several deputy chief physicians and attending physicians, and it scored 8 points. Hospital A: The cardiology department has complete equipment and facilities, including advanced examination equipment and treatment equipment, and it scored 10 points. Hospital B: The cardiology department has relatively backward equipment and facilities, and lacks advanced examination equipment, and it scored 4 points. Hospital C: The cardiology department has relatively complete equipment and facilities, but some of the equipment is slightly outdated, and it scored 8 points. Add up the scores of each hospital in the four aspects to get the basic score: Hospital A: 10 (geographical location) + 10 (professional characteristics) + 10 (doctor resources) + 10 (equipment and facilities) = 40 points; Hospital B: 8 (geographical location) + 0 (professional characteristics) + 2 (doctor resources) + 4 (equipment and facilities) = 14 points; Hospital C: 2 (geographical location) + 10 (professional characteristics) + 8 (doctor resources) + 8 (equipment and facilities) = 28 points.

[0043] The embodiment of the present application scores hospitals from multiple dimensions such as geographical location, professional characteristics, doctor resources and equipment and facilities to ensure the comprehensiveness and accuracy of the evaluation. The evaluation of geographical location takes into account the commuting time and commuting costs, reflecting the concern for user convenience; the evaluation of professional characteristics focuses on the matching degree between the hospital and the user's future medical needs to ensure the professionalism of the recommended hospital; the evaluation of doctor resources and equipment and facilities further ensures the quality and level of medical services in the hospital. The embodiment of the present application scores based on objective data (such as commuting time, commuting costs, doctor resource statistics, equipment and facility lists, etc.), avoiding subjective assumptions and prejudices, and improving the fairness and objectivity of the evaluation. The embodiment of the present application screens and scores hospitals according to the user's future medical needs and cost budget, and realizes personalized recommendations. By accurately grasping the needs of users, hospitals that better meet their needs can be recommended to users, improving the accuracy and satisfaction of recommendations. The embodiment of the present application quantitatively evaluates hospitals through a scoring system, which can quickly screen out hospitals that meet user needs and improve the efficiency of recommendations. This helps users quickly find hospitals that suit them and reduce selection difficulties and time costs. The embodiment of the present application provides users with a more convenient way to choose medical treatment through a scoring system. Users only need to enter their medical needs and cost budget to obtain a detailed hospital rating and recommendation list, which greatly saves users' time and energy.

[0044] Optionally, scoring the historical medical hospitals in the first hospital cluster according to the medical insurance usage data and the user's evaluation index to obtain additional points includes: Utilize natural language processing technology or machine learning algorithms to analyze the diagnostic information in the historical medical records, identify the type of disease the user suffers from and the severity of the disease, and determine the trend of disease changes based on physiological data at different time points in the historical medical records; determining a fifth score according to the disease type, the severity and the disease change trend, and determining a sixth score according to the user's evaluation index; The fifth score and the sixth score are added together to obtain the additional score.

[0045] Continuing with the above example, Mr. Zhang's medical history records contain multiple diagnosis information, such as "hypertension", "diabetes", "upper respiratory tract infection", etc. Using natural language processing technology or machine learning algorithms, it can be identified that Mr. Zhang suffers from multiple chronic diseases, and the description of the condition (such as "mild", "moderate", "severe") is also recorded at each visit. Mr. Zhang's medical history records also contain physiological data from multiple physical examinations, such as blood pressure, blood sugar, white blood cell count, etc. By analyzing the changing trends of these physiological data at different time points, it can be found that Mr. Zhang's blood pressure and blood sugar levels have increased in the past few years, and the white blood cell count will increase briefly every time he has an upper respiratory tract infection. Based on the type of disease Mr. Zhang suffers from and the severity of his condition, a base score can be set for him. For example, hypertension and diabetes, as chronic diseases, may be assigned a certain score respectively; and the severity of the condition (such as mild, moderate, severe) will affect the specific values ​​of these scores. Considering that Mr. Zhang's blood pressure and blood sugar levels are on the rise, this may mean that his health is gradually deteriorating. Therefore, the base score can be adjusted based on this trend to reflect the long-term changes in Mr. Zhang's condition. Add the scores of disease type, severity and disease change trend to get the fifth score. This score reflects the overall effect of Mr. Zhang's treatment in the historical hospitals and the changes in his health status. Mr. Zhang may have evaluated the hospitals he visited before, such as satisfaction scores, service evaluations, etc. The service quality and treatment effect of the hospital can be evaluated based on these evaluation indexes. According to the level of user evaluation index, a corresponding score can be set for each hospital. For example, a hospital with a higher satisfaction score may get a higher score; while a hospital with a poor service evaluation may get a lower score. Add the fifth score and the sixth score to get the additional score for each hospital. This additional score reflects the overall performance of the hospital in treating Mr. Zhang's disease, including treatment effect, service quality and other aspects.

[0046] By using natural language processing technology or machine learning algorithms, the diagnostic information in the historical medical records can be accurately analyzed to identify the type of disease and the severity of the disease suffered by the user. This accurate identification helps to more accurately evaluate the performance of the historical medical hospital in diagnosis and treatment, thereby improving the accuracy of the score. By analyzing the physiological data at different time points in the historical medical records, the changing trend of the disease can be determined. This helps to evaluate the hospital's ability in disease monitoring and management, as well as the hospital's response speed and effect to disease progression. Combined with the user's evaluation index (such as satisfaction, trust, etc.), a more personalized score can be given to the historical medical hospital. This personalized score not only takes into account the objective performance of the hospital, but also takes into account the user's subjective feelings, making the scoring result more comprehensive and accurate. The embodiment of the present application not only takes into account the objective performance of the hospital (such as disease diagnosis accuracy, disease change monitoring ability, etc.), but also takes into account the user's subjective evaluation (such as satisfaction, etc.). This helps to fully reflect the service quality of the hospital and provide the hospital with a direction for improvement and improvement. By scoring the historical medical hospital, the hospital can be encouraged to continuously improve the service quality and improve the medical level.

[0047] Optionally, the step of scoring all doctors in the second hospital that meet the medical field based on doctor grades, shift schedules, and patient evaluation indexes to determine target doctors includes: Determine a target medical field according to the future medical needs, and screen the doctors in the second hospital set according to the target medical field to obtain a set of doctors to be selected; The seventh score is determined based on the doctor's professional title and years of practice, the eighth score is determined based on the doctor's appointment availability and average waiting time, and the ninth score is determined based on the patient evaluation index; The seventh score, the eighth score and the ninth score are weightedly summed to obtain a doctor score, and the doctor with the highest doctor score is determined as the target doctor.

[0048] Continuing with the above example, Mr. Zhang's future medical needs are heart problems, so the target medical field is cardiac surgery. In the second hospital, all doctors with a professional background in cardiac surgery are screened to form a set of candidates. Assume that there are three doctors in the candidate set: Doctor A (Chief Physician, 20 years of practice), Doctor B (Deputy Chief Physician, 15 years of practice), and Doctor C (Attending Physician, 10 years of practice). According to the title and years of practice, a scoring standard can be set, such as chief physicians with more than 15 years of practice get the highest score, deputy chief physicians with 10-15 years of practice are second, and attending physicians with 5-10 years of practice are third. According to this standard, Doctor A may get the highest score of seventh. Assume that Doctor A's available appointment time is very tight, with an average waiting time of 3 weeks; Doctor B's available appointment time is relatively ample, with an average waiting time of 1 week; Doctor C's available appointment time is medium, with an average waiting time of 2 weeks. A scoring standard can be set, such as the shorter the waiting time, the higher the score. According to this standard, Doctor B may get the highest score of 8th. Suppose that according to the patient's evaluation index, Doctor A scored 4.8 points (out of 5 points), Doctor B scored 4.5 points, and Doctor C scored 4.7 points. The patient evaluation index can be directly used as the ninth score. Suppose that the doctor's level (seventh score) is the most important, with a weight of 0.5; the scheduling situation and average waiting time (eighth score) are second, with a weight of 0.3; the patient evaluation index (ninth score) is third, with a weight of 0.2. Calculate the total score of each doctor based on the set weights and scoring criteria. For example, Doctor A's total score = 0.5*Doctor A's seventh score + 0.3*Doctor A's eighth score + 0.2*Doctor A's ninth score. Suppose that after calculation, Doctor A has the highest total score, then Doctor A is determined to be Mr. Zhang's target doctor.

[0049] By determining the target medical field according to future medical needs and screening the doctors in the second hospital, the selection range can be quickly narrowed to ensure that the professional field of the selected doctor is highly matched with the patient's medical needs. This precise matching helps to improve the treatment effect and patient satisfaction. The embodiment of the present application not only takes into account the doctor's title and years of practice (seventh score), which usually reflect the doctor's professional level and experience, but also combines the doctor's appointment time and average waiting time (eighth score), which reflects the doctor's accessibility and service efficiency. At the same time, the introduction of the patient evaluation index (ninth score) makes the evaluation more comprehensive and can reflect the doctor's service quality and patient feedback. By combining the seventh score, the eighth score and the ninth score in a weighted summation method, a comprehensive doctor score is obtained, and the target doctor is determined accordingly, which can ensure that the selected doctor performs well in professional ability, service efficiency and service quality. This comprehensive consideration helps to improve the patient's medical experience, reduce waiting time, and improve treatment effect. The embodiment of the present application makes the selection of doctors more objective and fair by quantitatively evaluating multiple dimensions of doctors. This will help promote the rational allocation of medical resources, enable patients to more easily obtain high-quality medical services, and also motivate doctors to continuously improve their professional abilities and service quality.

[0050] Optionally, the method further includes: Determine the user's medication habits based on the medical insurance usage data, where the medication habits include drug purchase records, medication frequency, and medication cycle; Building a health profile of the user based on the medication habits and physical examination report; In combination with medical knowledge and health guidelines, health guidance suggestions are provided to the user based on the health portrait.

[0051] Extract the user's drug purchase records from the medical insurance usage data. This includes the name, quantity, purchase time and other information of all the drugs purchased by the user. Collect the user's medication frequency and medication cycle through user feedback or medical institution records. Medication frequency refers to the number of times a user takes a drug every day or week, and medication cycle refers to the length of time a user continues to take a certain drug. Sort and analyze the collected drug purchase records to identify the user's commonly used drug categories, brands and purchase frequency. Combined with medication frequency and medication cycle, assess whether the user's medication habits are stable and regular, and whether there are potential medication risks or dependence. Collect the user's physical examination report, including various physiological indicators (such as blood pressure, blood sugar, blood lipids, etc.), disease diagnosis results and doctor's advice. Sort and analyze the physical examination report to extract information related to the user's health status. Combine the user's medication habits and physical examination report to build the user's health profile. The health profile is a comprehensive overview of health information, including the user's physical condition, disease history, medication habits and other aspects. Through data mining and machine learning technology, further analyze and mine the user's health profile to discover potential health problems and risks. Collect and integrate medical knowledge and health guidelines, including information on disease prevention, healthy diet, exercise recommendations, etc. Provide users with personalized health guidance recommendations based on their health profiles, medical knowledge, and health guidelines. These recommendations may include prevention and treatment recommendations for specific diseases, healthy diet and exercise plans, and medication guidance. With the support of intelligent algorithms or expert systems, recommendations are made more accurate and in line with users' actual needs. Encourage users to manage their health and prevent diseases in accordance with the health guidance recommendations provided. Regularly track and evaluate users' health status, and adjust health guidance recommendations based on the evaluation results to ensure the effectiveness and applicability of the recommendations.

[0052] By analyzing the user's medication habits (including drug purchase records, medication frequency and medication cycle) and physical examination reports, a personalized health profile of the user can be constructed. This personalized health profile makes health guidance recommendations more in line with the user's actual situation and needs, and improves the pertinence and effectiveness of health guidance. Health guidance recommendations combine medical knowledge and health guidelines to ensure the scientificity and professionalism of the recommendations. Users can adjust their lifestyles, medication habits, etc. based on these recommendations, thereby improving their own health management level and preventing the occurrence or deterioration of diseases. By providing personalized health guidance recommendations, users can reduce the situation where they blindly seek medical treatment or abuse drugs due to lack of understanding of their own health status. This helps promote the rational use of medical resources and reduce the burden on the medical system. The provision of health guidance recommendations can enhance users' health awareness and make users pay more attention to their own health status.

[0053] Optionally, the health guidance suggestions provided to the user based on the health portrait in combination with medical knowledge and health guidelines include: Assess the disease risk of the user according to the health profile, and determine the target disease whose disease risk is higher than a threshold; Personalized dietary advice is provided based on the influencing factors of the target disease and the physiological indicators of the user, wherein the dietary advice includes recommended food types, intake amounts, and cooking methods.

[0054] In Mr. Zhang's health portrait, it was found that his blood pressure was high and he frequently bought antihypertensive drugs in recent years. At the same time, his physical examination report also showed that his blood lipid level had increased. According to medical knowledge and health guidelines, it can be known that high blood pressure and high blood lipids are important risk factors for cardiovascular disease. Therefore, Mr. Zhang was assessed to have a higher risk of cardiovascular disease than the general population and was identified as a target disease. In view of the influencing factors of cardiovascular disease, such as high-salt and high-fat diets, the following personalized dietary recommendations were provided to Mr. Zhang: Recommended food types: Mr. Zhang is advised to eat more foods rich in potassium, magnesium, calcium and other trace elements, such as kelp, seaweed, celery, spinach, peanuts, walnuts, etc. These foods help lower blood pressure and blood lipids. At the same time, he is recommended to eat a moderate amount of foods rich in high-quality protein, such as lean meat, fish, beans, etc., to meet the body's basic nutritional needs; Intake: It is recommended that Mr. Zhang control his total daily calorie intake and reduce the intake of high-fat and high-sugar foods, such as fried foods and sweets. At the same time, control the intake of salt to no more than 6 grams per day to lower blood pressure levels; Cooking methods: Mr. Zhang is advised to use healthy cooking methods such as steaming, boiling, stewing, and baking, and avoid high-oil cooking methods such as frying and stir-frying. At the same time, he is encouraged to use vegetable oil for cooking and control his daily vegetable oil intake.

[0055] The embodiment of the present application combines medical knowledge and health guidelines, and uses multi-dimensional data in health portraits (such as medical insurance usage data, medication habits, physical examination reports, etc.) to assess the user's risk of illness. This evaluation method has a scientific basis and can more accurately reflect the user's health status. Through in-depth analysis of health portraits, it is possible to promptly discover potential health risks of users, providing strong support for early prevention and treatment. Provide personalized dietary advice based on the influencing factors of the target disease and the user's physiological indicators. This advice not only takes into account the specificity of the disease, but also combines the individual differences of the user, so it is highly targeted. Dietary advice includes recommended food types, intake, and cooking methods, which are all based on medical knowledge and health guidelines, so they are scientific and reasonable. The dietary advice provided is specific and clear, and users can easily understand and implement it. This helps users better adjust their eating habits, thereby reducing the risk of illness.

[0056] This embodiment also discloses a personalized recommendation system for medical services. Figure 2 is a schematic diagram of a module of a personalized recommendation system for medical services disclosed in an embodiment of the present application, such as Figure 2 As shown, the system includes a collection module 201, a feature module 202, a prediction module 203 and a recommendation module 204, wherein: The collection module 201 is configured to obtain the user's medical insurance usage data, wherein the medical insurance usage data includes historical medical records, fee settlement records, drug purchase records and physical examination reports; A feature module 202 is configured to extract target features associated with medical insurance usage habits and needs from the medical insurance usage data, wherein the target features include medical frequency, medical type, expense type, disease type, and medication habits; Prediction module 203, configured to use a preset deep learning model to predict the user's future medical needs and cost budget according to the target features; The recommendation module 204 is configured to recommend a target hospital and a target doctor according to the future medical needs and cost budget.

[0057] Optionally, the recommendation module 204 is configured to: According to the future medical needs and cost budget, all hospitals in the current area are screened to obtain a first hospital set; Score all hospitals in the first hospital cluster according to their geographical location, professional characteristics, doctor resources, and equipment and facilities to obtain a basic score; Determine a historical medical hospital according to the historical medical records, and score the historical medical hospitals concentrated in the first hospital according to the medical insurance usage data and the user's evaluation to obtain additional points; Obtaining a total hospital score according to the basic score and the additional score, and screening the first hospital set according to the total hospital score to obtain a second hospital set; All doctors in the second hospital that meet the medical field requirements are scored based on doctor grade, scheduling and patient evaluation index to determine the target doctor.

[0058] Optionally, the recommendation module 204 is configured to: The first score is determined based on the geographical location of the hospital and the commuting time and commuting cost of the target location preset by the user, and the second score is determined based on the matching degree between the hospital's specialty departments and the user's future medical needs; Determine a target department according to the future medical needs, determine a third score according to the richness of the doctor resources of the target department of the hospital, and determine a fourth score according to the perfection of the equipment and facilities associated with the target department in the hospital; The first score, the second score, the third score, and the fourth score are added together to obtain the basic score.

[0059] Optionally, the recommendation module 204 is configured to: Utilize natural language processing technology or machine learning algorithms to analyze the diagnostic information in the historical medical records, identify the type of disease the user suffers from and the severity of the disease, and determine the trend of disease changes based on physiological data at different time points in the historical medical records; determining a fifth score according to the disease type, the severity and the disease change trend, and determining a sixth score according to the user's evaluation index; The fifth score and the sixth score are added together to obtain the additional score.

[0060] Optionally, the recommendation module 204 is configured to: Determine a target medical field according to the future medical needs, and screen the doctors in the second hospital set according to the target medical field to obtain a set of doctors to be selected; The seventh score is determined based on the doctor's professional title and years of practice, the eighth score is determined based on the doctor's appointment availability and average waiting time, and the ninth score is determined based on the patient evaluation index; The seventh score, the eighth score and the ninth score are weightedly summed to obtain a doctor score, and the doctor with the highest doctor score is determined as the target doctor.

[0061] Optionally, the system further comprises a suggestion module, wherein the suggestion module is configured to: Determine the user's medication habits based on the medical insurance usage data, where the medication habits include drug purchase records, medication frequency, and medication cycle; Building a health profile of the user based on the medication habits and physical examination report; In combination with medical knowledge and health guidelines, health guidance suggestions are provided to the user based on the health portrait.

[0062] Optionally, the suggestion module is configured to: Assess the disease risk of the user according to the health profile, and determine the target disease whose disease risk is higher than a threshold; Personalized dietary advice is provided based on the influencing factors of the target disease and the physiological indicators of the user, wherein the dietary advice includes recommended food types, intake amounts, and cooking methods.

[0063] It should be noted that: when the device provided in the above embodiment realizes its function, only the division of the above functional modules is used as an example. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0064] This embodiment also discloses an electronic device, referring to Figure 3 The electronic device may include: at least one processor 301 , at least one communication bus 302 , a user interface 303 , a network interface 304 , and at least one memory 305 .

[0065] The communication bus 302 is used to realize the connection and communication between these components.

[0066] The user interface 303 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 303 may also include a standard wired interface and a wireless interface.

[0067] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0068] Among them, the processor 301 may include one or more processing cores. The processor 301 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 305, and calling data stored in the memory 305. Optionally, the processor 301 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 301 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 301, and it can be implemented separately through a chip.

[0069] Among them, the memory 305 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 305 includes a non-transitory computer-readable storage medium. The memory 305 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 305 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 305 may also be optionally at least one storage device located away from the aforementioned processor 301. As Figure 3 As shown, the memory 305 as a computer storage medium may include an operating system, a network communication module, a user interface module, and an application program for a personalized recommendation method for medical services.

[0070] exist Figure 3 In the electronic device shown, the user interface 303 is mainly used to provide an input interface for the user and obtain data input by the user; and the processor 301 can be used to call the application program storing the personalized recommendation method for medical services in the memory 305. When executed by one or more processors 301, the electronic device executes one or more methods in the above-mentioned embodiments.

[0071] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the order of the actions described, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0072] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0073] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which 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 integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0074] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, 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 solution of this embodiment.

[0075] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0076] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory 305 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 steps of the various embodiments of the present application. The aforementioned memory 305 includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk.

[0077] The above is only an exemplary embodiment of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification, those skilled in the art will easily think of other embodiments of the present disclosure. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the technical field that are not recorded in the present disclosure. The description and examples are only regarded as exemplary, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A personalized recommendation method for medical services, characterized in that: Applied to the medical insurance service platform, the method includes: Obtain the user's medical insurance usage data, which includes historical medical records, fee settlement records, drug purchase records, and physical examination reports; Extracting target features associated with medical insurance usage habits and needs from the medical insurance usage data, the target features including medical frequency, medical type, expense type, disease type, and medication habits; Using a preset deep learning model to predict the user's future medical needs and cost budget according to the target features; Based on the future medical needs and cost budget, recommend target hospitals and target doctors.

2. The personalized recommendation method for medical services according to claim 1, characterized in that: According to the future medical needs and cost budget, the recommended target hospitals and target doctors include: According to the future medical needs and cost budget, all hospitals in the current area are screened to obtain a first hospital set; Score all hospitals in the first hospital cluster according to their geographical location, professional characteristics, doctor resources, and equipment and facilities to obtain a basic score; Determine a historical medical hospital according to the historical medical records, and score the historical medical hospitals concentrated in the first hospital according to the medical insurance usage data and the user's evaluation to obtain additional points; Obtaining a total hospital score according to the basic score and the additional score, and screening the first hospital set according to the total hospital score to obtain a second hospital set; All doctors in the second hospital that meet the medical field requirements are scored based on doctor grade, scheduling and patient evaluation index to determine the target doctor.

3. The personalized recommendation method for medical services according to claim 2, characterized in that: The scoring of all hospitals in the first hospital cluster according to the hospital's geographical location, professional characteristics, doctor resources and equipment facilities to obtain the basic score includes: The first score is determined based on the geographical location of the hospital and the commuting time and commuting cost of the target location preset by the user, and the second score is determined based on the matching degree between the hospital's specialty departments and the user's future medical needs; Determine a target department according to the future medical needs, determine a third score according to the richness of the doctor resources of the target department of the hospital, and determine a fourth score according to the perfection of the equipment and facilities associated with the target department in the hospital; The first score, the second score, the third score, and the fourth score are added together to obtain the basic score.

4. The personalized recommendation method for medical services according to claim 2, characterized in that: Scoring the historical medical hospitals concentrated in the first hospital according to the medical insurance usage data and the user's evaluation index to obtain additional points includes: Utilize natural language processing technology or machine learning algorithms to analyze the diagnostic information in the historical medical records, identify the type of disease the user suffers from and the severity of the disease, and determine the trend of disease changes based on physiological data at different time points in the historical medical records; determining a fifth score according to the disease type, the severity and the disease change trend, and determining a sixth score according to the user's evaluation index; The fifth score and the sixth score are added together to obtain the additional score.

5. The personalized recommendation method for medical services according to claim 2, characterized in that: The target doctors are scored according to the doctor's grade, shift schedule and patient evaluation index to determine the target doctors: Determine a target medical field according to the future medical needs, and screen the doctors in the second hospital set according to the target medical field to obtain a set of doctors to be selected; The seventh score is determined based on the doctor's professional title and years of practice, the eighth score is determined based on the doctor's appointment availability and average waiting time, and the ninth score is determined based on the patient evaluation index; The seventh score, the eighth score and the ninth score are weightedly summed to obtain a doctor score, and the doctor with the highest doctor score is determined as the target doctor.

6. The personalized recommendation method for medical services according to claim 1, characterized in that: The method further comprises: Determine the user's medication habits based on the medical insurance usage data, where the medication habits include drug purchase records, medication frequency, and medication cycle; Building a health profile of the user based on the medication habits and physical examination report; In combination with medical knowledge and health guidelines, health guidance suggestions are provided to the user based on the health portrait.

7. The personalized recommendation method for medical services according to claim 6, characterized in that: The health guidance suggestions provided to the user based on the health portrait in combination with medical knowledge and health guidelines include: Assess the disease risk of the user according to the health profile, and determine the target disease whose disease risk is higher than a threshold; Personalized dietary advice is provided based on the influencing factors of the target disease and the physiological indicators of the user, wherein the dietary advice includes recommended food types, intake amounts, and cooking methods.

8. A personalized recommendation system for medical services, characterized in that: It includes acquisition module, feature module, prediction module and recommendation module, among which: A collection module configured to obtain the user's medical insurance usage data, wherein the medical insurance usage data includes historical medical records, fee settlement records, drug purchase records, and physical examination reports; a feature module configured to extract target features associated with medical insurance usage habits and needs from the medical insurance usage data, the target features including medical frequency, medical type, expense type, disease type, and medication habits; A prediction module, configured to use a preset deep learning model to predict the user's future medical needs and cost budget according to the target features; The recommendation module is configured to recommend a target hospital and a target doctor based on the future medical needs and cost budget.

9. An electronic device, characterized in that: It includes a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed, the method according to any one of claims 1 to 7 is performed.

Images