Intelligent medical online service system and online service method

Abstract

The invention provides an intelligent medical online service system and an online service method. The system comprises a service platform, terminal equipment, a community service station and a hospital service terminal, the service platform is connected with a community service station and a hospital service terminal; the terminal equipment is used for a user to register and access the service platform; the service platform comprises a registration and login module, a self-service module, a doctor-seeing appointment module, a doctor-seeing statistics module, a service appointment module, an evaluation module and an optimization module, a user accesses the platform through the terminal equipment, obtains services through different modules, completes inquiry and evaluation, gives preliminary auxiliary diagnosis suggestions through the platform, and optimizes the diagnosis through the terminal equipment. The user is helped to obtain some reference suggestions before appointment to see a doctor, so that the user can better understand the own illness state; the system can reasonably allocate resources of doctors and treatment equipment, ensures effective utilization of the resources, and improves the accessibility of treatment service; and the system performance and the user experience are improved through continuous optimization.

Description

Technical Field

[0001] This invention relates to the field of medical service technology, and in particular to a smart medical online service system and online service method. Background Technology

[0002] Due to the imperfections in China's public healthcare management system, high medical costs, limited access, and low coverage are problems plaguing the public. In particular, the "inefficient healthcare system, substandard medical services, and the difficulty and expense of accessing medical care" are major focuses of social concern. Overcrowded large hospitals, underutilized community clinics, and cumbersome patient procedures are all problems stemming from poor access to medical information, the polarization of medical resources, and low healthcare efficiency. These issues have become significant factors affecting harmonious social development. Summary of the Invention

[0003] This invention provides a smart healthcare online service system and method to improve the medical experience, medical efficiency, and optimize resource allocation.

[0004] This invention proposes a smart healthcare online service system, the system comprising:

[0005] Service platforms, terminal equipment, community service stations, and hospital service terminals;

[0006] The service platform connects community service stations and hospital service terminals;

[0007] The terminal device is used for user registration and access to the service platform;

[0008] The service platform includes a registration and login module, a self-service module, an appointment module, an appointment statistics module, an appointment service module, an evaluation module, and an optimization module.

[0009] Furthermore, in a smart healthcare online service system, the registration and login module includes:

[0010] Users register account information through the registration module. After registration, they can select to add patient information, which includes name, gender, age, contact person, address, ID number, and emergency contact person. Based on the patient's ID number, the patient's social security account is directly linked.

[0011] Establish health records for patients.

[0012] Furthermore, in a smart healthcare online service system, the self-service module's method includes:

[0013] Obtain the patient's medical history and past medical records;

[0014] Based on the description of the illness and past medical records, the platform intelligently provides preliminary auxiliary diagnostic suggestions;

[0015] Based on the aforementioned auxiliary suggestions, patients may choose to purchase medication directly, consult a doctor, or undergo observation; the description of the condition includes voice description, text description, photos, and / or examination reports;

[0016] Based on the description of the illness and past medical records, the platform intelligently provides preliminary auxiliary diagnostic suggestions;

[0017] The collected descriptions of the patient's condition and past medical records are stored in a structured format in the database;

[0018] Create corresponding tables in the database, including a patient information table, a disease description table, and a treatment record table, and link the tables through primary key-foreign key relationships;

[0019] Data preprocessing of disease descriptions and past medical records yields preprocessed information, including text cleaning and standardization;

[0020] Natural Language Processing (NLP) technology is used to transform preprocessed text information into vector representations and extract useful features; the useful features include symptoms, keywords, and medical history information.

[0021] Extract numerical data from medical records;

[0022] Cluster analysis was performed on a large number of cases on the platform, and combined with doctors' diagnostic results, to obtain a set of cluster features for each disease;

[0023] The newly acquired disease features are compared with the features in the cluster feature set to obtain the similarity S. yi ;

[0024]

[0025] Where w1, w2...wn are the weights of the features, S1, S2...Sn are the similarities of each feature, and n is the number of features included; n = min(P, Q), where P is the newly acquired disease feature, and Q is the number of features corresponding to a certain disease in the feature library;

[0026] For different disease feature databases, different similarity scores are obtained, and the top three similarity scores S are... y1 S y2 S y3 ;

[0027] If (S) y1 -S y2 ) / S y1 ≥50%; then S y1The corresponding disease is given suggestions as a preliminary auxiliary diagnosis;

[0028] If (S) y1 -S y2 ) / S y1 <50%; and (S y2 -S y3 ) / S y3 ≥50%; then S y1 and S y2 The corresponding diseases are given suggestions as preliminary auxiliary diagnoses; and the probabilities are compared based on the degree of similarity.

[0029] If (S) y1 -S y2 ) / S y1 <50%; and (S y2 -S y3 ) / S y3 <50%; then S y1 S y2 S y3 The corresponding disease is given as a preliminary auxiliary diagnosis, and the probability comparison results are given based on the similarity.

[0030] Furthermore, in a smart healthcare online service system, the appointment booking module includes:

[0031] Offers the option of online or offline appointment booking;

[0032] If a patient chooses online consultation, the system sets the consultation duration based on the patient's description of their condition and past medical records; the consultation duration is:

[0033]

[0034] Where L represents the network speed during the consultation, L a Let n be the average network speed, n be the number of patients a doctor sees simultaneously, and v be the average network speed. a Average doctor speed, v is the doctor's response speed, and T0 is the preset time;

[0035] Where k is a coefficient, obtained from the platform's preliminary auxiliary diagnosis, if (S y1 -S y2 ) / S y1 ≥50%; then k=1;

[0036] If (S) y1 -S y2 ) / S y1 <50%; and (S y2 -S y3 ) / S y3≥50%; then k=1+(S y1 -S y2 ) / S y1 ;

[0037] If (S) y1 -S y2 ) / S y1 <50%; and (S y2 -S y3 ) / S y3 <50%; then k = 1 + (S y1 +S y2 ) / (S y1 +S y2 +S y3 );

[0038] Offline appointment booking provides a demand selection window, which includes a first choice and a second choice;

[0039] The first option includes specialist appointments and regular medical visits;

[0040] The second option is the selection of a doctor, including the doctor's gender and preference based on experience, attitude, or overall score;

[0041] If a patient chooses a regular appointment, the system will automatically match the corresponding department and doctor based on the patient's second choice and the number of registered patients, and generate a queue number; the queue number information will be updated in real time and a reminder will be sent.

[0042] If a patient chooses to make an appointment with a specialist, a list of specialists, their introductions, and their schedules will be provided based on the hospital department the patient selects. The patient can then choose a suitable specialist and appointment time from the list.

[0043] Furthermore, in a smart healthcare online service system, the patient visit statistics module includes:

[0044] Historical medical records statistics and appointment statistics;

[0045] Historical records include:

[0046] The number of patients visiting each department within each time period was counted by dividing the time period into time periods; online and offline visits were counted separately.

[0047] The day is divided into time periods by hour, and the number of patients in each department during each time period is counted.

[0048] The average number of patients visited by each department during each time period within a week is calculated; statistics are compiled separately for rest days and workdays;

[0049] By using the statistical data of each month over the past year as an observation window, the distribution of the number of patients in each department can be obtained, including peak time periods, off-peak time periods, and low time periods.

[0050] Based on the distribution, resources should be allocated reasonably, including arranging the working hours and number of doctors and nurses, considering increasing outpatient hours, arranging expert consultations, or opening evening outpatient clinics for specific departments.

[0051] Furthermore, in a smart healthcare online service system, the appointment service module includes:

[0052] Schedule VR online guidance services and in-home services;

[0053] For in-home services, the appointment service module obtains the patient's address or location information and the services requested. Based on the address or location information provided by the patient and the services requested, it displays information about doctors or nurses at nearby community hospitals, including their names, titles, and areas of expertise. The patient then makes an appointment based on this information.

[0054] Furthermore, in a smart healthcare online service system, the evaluation module includes:

[0055] System evaluation feedback, doctor evaluation, and home visit service evaluation;

[0056] The system evaluation feedback includes evaluation of system response speed, usability, error feedback, and improvement suggestions;

[0057] The doctor evaluation includes evaluations of online consultations and offline visits, and the evaluation includes evaluations of professional competence, communication skills, service attitude, and response speed.

[0058] Furthermore, in a smart healthcare online service system, the optimization module includes:

[0059] System function optimization, service optimization, and model optimization;

[0060] The service optimization includes personnel training;

[0061] The model optimization includes:

[0062] The accuracy of model-assisted diagnosis is obtained by comparing the doctor's diagnosis results with the auxiliary diagnosis suggestions provided by the platform.

[0063]

[0064] Where k is a coefficient, obtained from the platform's preliminary auxiliary diagnosis, F Mj For primary diagnostic similarity; F CjFor secondary diagnostic similarity, F Zj The similarity of treatment recommendations; m is the number of cases;

[0065] Optimize the model regularly.

[0066] This invention proposes a smart healthcare online service method, the method comprising:

[0067] S1. Obtain the patient's medical condition description and past medical records; automatically generate the patient's health record based on the medical condition description and past medical records; and provide self-service, intelligently providing preliminary auxiliary diagnostic suggestions;

[0068] S2. Provide online appointment for medical treatment through online medical services, and provide offline appointment for medical treatment through offline appointment services;

[0069] S3. The system uses the medical visit statistics module to collect historical medical visit records and the number of appointments. Based on the results of the historical medical visit records, resources are allocated reasonably. Based on the results of the appointments, user needs, and doctor scheduling, doctors are automatically matched.

[0070] S4. Provide appointment-based VR online guidance and on-site services through the appointment service module;

[0071] S5. Obtain system evaluation feedback, doctor evaluation, and home service evaluation through the evaluation module;

[0072] S6. Optimize system functions, services, and models through the optimization module.

[0073] Furthermore, in a smart healthcare online service method, step S1 includes:

[0074] The collected descriptions of the patient's condition and past medical records are stored in a structured format in the database;

[0075] Create corresponding tables in the database, including a patient information table, a disease description table, and a treatment record table, and link the tables through primary key-foreign key relationships;

[0076] Data preprocessing of disease descriptions and past medical records yields preprocessed information, including text cleaning and standardization;

[0077] Natural Language Processing (NLP) technology is used to transform preprocessed text information into vector representations and extract useful features; the useful features include symptoms, keywords, and medical history information.

[0078] Extract numerical data from medical records;

[0079] Cluster analysis was performed on a large number of cases on the platform, and combined with doctors' diagnostic results, to obtain a set of cluster features for each disease;

[0080] The newly acquired disease features are compared with the features in the cluster feature set to obtain the similarity S. yi ;

[0081]

[0082] Where w1, w2…wn are the weights of the features, S1, S2…Sn are the similarities of each feature, and n is the number of features included; n = min(P, Q), where P is the newly acquired disease feature, and Q is the number of features corresponding to a certain disease in the feature library;

[0083] For different disease feature databases, different similarity scores are obtained, and the top three similarity scores S are... y1 S y2 S y3 ;

[0084] If (S) y1 -S y2 ) / S y1 ≥50%; then S y1 The corresponding disease is given suggestions as a preliminary auxiliary diagnosis;

[0085] If (S) y1 -S y2 ) / S y1 <50%; and (S y2 -S y3 ) / S y3 ≥50%; then S y1 and S y2 The corresponding diseases are given suggestions as preliminary auxiliary diagnoses; and the probabilities are compared based on the degree of similarity.

[0086] If (S) y1 -S y2 ) / S y1 <50%; and (S y2 -S y3 ) / S y3 <50%; then S y1 S y2 S y3 The corresponding disease is given as a preliminary auxiliary diagnosis, and the probability comparison results are given based on the similarity.

[0087] The beneficial effects of this invention are as follows: Through the intelligent medical online service system and method described in this invention, users can choose to receive medical treatment through online or offline appointments, eliminating the need for traditional queuing. This significantly reduces patient waiting time and improves efficiency. The system generates health records based on the patient's condition description and past medical records, and automatically provides preliminary auxiliary diagnostic suggestions. This helps users obtain reference opinions before making an appointment and better understand their condition. By statistically analyzing historical medical records and the number of appointments, the system can rationally allocate resources for doctors and medical equipment, ensuring effective resource utilization and improving the accessibility of medical services. The system automatically matches suitable doctors based on the number of appointments and user needs. Furthermore, the appointment service module provides VR online guidance and home visit services, offering users a more convenient way to receive medical treatment. The evaluation module obtains user feedback on the system, doctors, and home visit services. This helps understand user satisfaction and areas for improvement, optimize system functions and service processes, and enhance user experience. The system optimizes its functions, services, and models through the optimization module and continuously improves based on user feedback. Continuous optimization can improve system performance and user experience, providing more accurate and efficient auxiliary diagnostic services. This smart healthcare online service approach can provide convenient access to medical care, auxiliary diagnostic suggestions, rational allocation of resources, doctor matching and guidance services, and improve system performance and user experience through evaluation feedback and continuous optimization. Attached Figure Description

[0088] Figure 1 This is a schematic diagram of an intelligent medical online service system according to the present invention. Detailed Implementation

[0089] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0090] Numerous specific details are set forth in the following description to provide a thorough understanding of the invention. The described embodiments are only a part of, and not all, of the embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0091] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.

[0092] This embodiment discloses a smart healthcare online service system, the system comprising:

[0093] Service platforms, terminal equipment, community service stations, and hospital service terminals;

[0094] The service platform connects community service stations and hospital service terminals;

[0095] The terminal device is used for user registration and access to the service platform;

[0096] The service platform includes a registration and login module, a self-service module, an appointment module, a medical statistics module, an appointment service module, an evaluation module, and an optimization module.

[0097] The registration and login module includes:

[0098] Users register account information through the registration module. After registration, they can select to add patient information, which includes name, gender, age, contact person, address, ID number, and emergency contact person. Based on the patient's ID number, the patient's social security account is directly linked.

[0099] Establish health records for patients.

[0100] The working principle of the above technical solution is as follows:

[0101] Users register by filling in their account information through the registration module and then select to add patient information. Patient information includes name, gender, age, contact person, address, ID number, and emergency contact person.

[0102] The system directly links to the patient's social security account based on their ID card number, enabling subsequent business operations using the social security account. It also establishes a health record for each patient to store and manage their personal health information. Users can access the service platform using terminal devices and perform operations through the self-service module, such as querying personal health records, understanding medical service processes, and obtaining preliminary diagnostic suggestions from the platform. Users can select hospitals and departments, and schedule appointments with doctors through the appointment booking module. After a successful booking, users can view and manage their appointment information in the appointment service module, including appointments pending, already booked, and canceled. After receiving medical services, users can evaluate doctors and hospitals through the evaluation module, providing feedback and suggestions. This helps other users understand the service quality of doctors and hospitals and improve the level of medical services. The system continuously collects user feedback and evaluations through the optimization module, conducting data analysis and mining to further optimize the service platform's functionality and user experience.

[0103] The above technical solution achieves the following results: Through the coordination and collaborative work of the various modules, this smart healthcare online service system enables functions such as user registration and login, self-service, appointment scheduling, evaluation, and optimization. Users can conveniently manage their personal health records, schedule appointments, evaluate doctors and hospitals, and continuously improve service quality through system optimization, providing users with a better smart healthcare online service experience.

[0104] This embodiment discloses a smart healthcare online service system, wherein the self-service module includes the following method:

[0105] Obtain the patient's medical history and past medical records;

[0106] Based on the description of the condition, the platform intelligently provides preliminary auxiliary diagnostic suggestions; the patient can choose to purchase medication directly, consult a doctor, or undergo observation based on the auxiliary suggestions; the description of the condition includes voice description, text description, photos and / or examination reports;

[0107] Voice description: Patients can use the voice input function provided by the smart healthcare platform to briefly describe their symptoms and discomfort. For example, patients can describe the location of their pain, the intensity of the pain, accompanying symptoms, as well as the duration and frequency of the symptoms.

[0108] Text description: Patients can describe their condition in detail in text form on the smart healthcare platform. They can write down the onset time of symptoms, their progression, triggers, factors that alleviate or aggravate the symptoms, etc. In addition, patients can also describe their past treatment history, medications, and any other relevant information related to the symptoms.

[0109] Photos: Patients can upload photos of relevant areas through the smart healthcare platform, which is very helpful for the diagnosis of skin diseases, trauma, or other visible symptoms; doctors can observe the photos to understand the nature, color, size, and distribution of lesions;

[0110] Medical Records: Patients can upload their past medical records (such as blood pressure and blood sugar data) to the smart healthcare platform. These records can provide more comprehensive information, helping doctors analyze the patient's health status and provide corresponding auxiliary diagnostic suggestions.

[0111] The collected descriptions of the patient's condition and past medical records are stored in a structured format in a database; a database management system (such as MySQL or MongoDB) can be used to manage the data.

[0112] Create corresponding tables in the database, including a patient information table, a disease description table, and a treatment record table, and link the tables through primary key-foreign key relationships;

[0113] Data preprocessing of disease descriptions and past medical records yields preprocessed information, including text cleaning and standardization;

[0114] Natural Language Processing (NLP) techniques are used to transform preprocessed text information into vector representations and extract useful features, including symptoms, keywords, and medical history information, such as using the bag-of-words model or Word2Vec.

[0115] Extract numerical data from medical records;

[0116] If the medical records contain time series data, use time series analysis methods to extract relevant features, such as statistical features, frequency domain features, and time domain features.

[0117] Cluster analysis was performed on a large number of cases on the platform, and combined with doctors' diagnostic results, to obtain a set of cluster features for each disease;

[0118] The newly acquired disease features are compared with the features in the cluster feature set to obtain the similarity S. yi ;

[0119]

[0120] Where w1, w2…wn are the weights of the features, S1, S2…Sn are the similarities of each feature, and n is the number of features included; n = min(P, Q), where P is the newly acquired disease feature, and Q is the number of features corresponding to a certain disease in the feature library; the weights of the features are derived from clinical experience, statistical analysis, machine learning methods, and expert consensus.

[0121] For different disease feature databases, different similarity scores are obtained, and the top three similarity scores S are... y1 S y2 S y3 ;

[0122] If (S) y1 -S y2 ) / S y1 ≥50%; then S y1 The corresponding disease is given suggestions as a preliminary auxiliary diagnosis;

[0123] If (S) y1 -S y2 ) / S y1 <50%; and (S y2 -S y3 ) / S y3 ≥50%; then S y1 and S y2 The corresponding diseases are given suggestions as preliminary auxiliary diagnoses; and the probabilities are compared based on the degree of similarity.

[0124] If (S) y1 -S y2 ) / S y1 <50%; and (S y2 -S y3 ) / S y3 <50%; then S y1 S y2 S y3 The corresponding disease is given as a preliminary auxiliary diagnosis, and the probability comparison results are given based on the similarity.

[0125] The working principle of the above technical solution is as follows: Acquire the patient's medical condition description and past medical records, including voice descriptions, text descriptions, photos, and / or examination reports; store the collected medical condition descriptions and past medical records in a structured form in a database, establishing corresponding patient information tables, medical condition description tables, and medical record tables, and linking them through primary key-foreign key relationships; preprocess the data in the medical condition description and past medical record tables, including text cleaning and standardization, to ensure data quality and consistency; use Natural Language Processing (NLP) technology to convert the preprocessed text information into vector representations and extract useful features, such as symptoms, keywords, and medical history information. Feature extraction can be performed using methods such as the bag-of-words model and Word2Vec. Numerical data from medical records is extracted, and time-series data is analyzed using time-series analysis methods to extract corresponding features, such as statistical features, frequency domain features, and time domain features. Cluster analysis is performed on a large number of case features on the platform, combined with doctors' diagnostic results, to obtain a cluster feature set for each disease. The newly acquired disease features are compared with the features in the cluster feature set to calculate similarity. Based on the similarity, preliminary auxiliary diagnostic suggestions are given. Depending on the situation, one disease may be suggested, or multiple diseases may be suggested, with a comparison of probabilities based on similarity. Feature weights can be determined based on clinical experience, statistical analysis, machine learning methods, and expert consensus to ensure the accuracy and reliability of the auxiliary diagnosis.

[0126] The effects of the above technical solution are as follows: By combining the patient's description of their condition and past medical records with the platform's intelligent preliminary auxiliary diagnostic suggestions, the accuracy of diagnosis can be improved; the system uses natural language processing technology to convert text information into vector representations and extract useful features, helping doctors better understand the patient's symptoms and disease progression; patients can conduct self-diagnosis anytime, anywhere through the smart healthcare platform, without waiting for appointments or queuing; it reduces resource waste for some mild cases, allowing patients to describe their condition in detail through voice, text, and uploaded photos, providing doctors with more comprehensive information and thus improving diagnostic efficiency; based on the patient's description of their condition and past medical records, the system will provide preliminary auxiliary diagnostic suggestions based on cluster analysis and similarity calculations; based on the similarity score, the system will provide a probability comparison result according to certain rules, helping patients understand various possible diseases and make appropriate choices; the system stores the collected descriptions of conditions and past medical records in a structured form in a database and manages them through a database management system. This facilitates subsequent data analysis and research, helps discover potential disease patterns, and provides more accurate diagnostic suggestions. The weight parameters in the formula can be adjusted based on clinical experience, statistical analysis, machine learning methods, and expert consensus to reflect the importance of different features for diagnosis. By setting appropriate weights, the similarity calculation can be made more accurate and reliable. The weight parameters in the formula can be adjusted according to the importance of different features, thus taking multiple feature factors into account and improving the comprehensiveness and accuracy of the diagnosis. Similarity is an indicator that measures the degree of similarity between newly acquired disease features and a set of features from a certain disease cluster. By calculating similarity, the degree of matching between new features and existing disease features can be assessed.

[0127] P represents the number of newly acquired disease features, and Q represents the number of features in the feature set for a certain disease. By comparing the magnitudes of P and Q, the number of features used for similarity calculation is determined to ensure an appropriate number, avoiding the influence of too many or too few features on the results. The feature weights (w1, w2, ..., wn) represent the importance of each feature in the similarity calculation. Weight values ​​can be determined by combining clinical experience, statistical analysis, machine learning methods, and expert consensus. A higher weight means that the feature has a greater influence in the similarity calculation. Based on the magnitude of similarity, the diseases are ranked, and the three most similar diseases are selected as candidates. These similarities can serve as preliminary auxiliary diagnostic suggestions. If the first-ranked similarity is significantly greater than the second-ranked similarity, it indicates a high correlation with the corresponding disease, while other similarities have low correlation; in this case, the first-ranked similarity is used as the preliminary auxiliary diagnostic suggestion. If the first and second-ranked similarities are not significantly different, both possibilities are relatively high; therefore, diseases corresponding to both similarities are used as preliminary auxiliary diagnostic suggestions, ranked by similarity. If the first, second, and third-ranked similarities are not significantly different, diseases corresponding to all three similarities are used as preliminary auxiliary diagnostic suggestions, ranked by similarity. Based on the above formula, similarity can be calculated between newly acquired disease features and existing cluster feature sets. Preliminary auxiliary diagnostic suggestions can be given based on the magnitude of the similarity and the comparison of differences. This helps doctors identify possible diseases more quickly and accurately, and provide corresponding diagnostic suggestions and probability comparison results.

[0128] In summary, the self-service module of the smart healthcare online service system can provide fast, accurate, and personalized diagnostic suggestions, allowing patients to access medical services anytime, anywhere, and improving the efficiency and accuracy of doctors' diagnoses. Furthermore, through data management and analysis, it can also provide valuable data resources for medical research.

[0129] This embodiment discloses a smart medical online service system, wherein the appointment module includes:

[0130] Offers the option of online or offline appointment booking;

[0131] If a patient chooses online consultation, the system sets the consultation duration based on the patient's description of their condition and past medical records; among which...

[0132] The consultation duration is:

[0133]

[0134] Where L represents the network speed during the consultation, L a Let n be the average network speed, n be the number of patients a doctor sees simultaneously, and v be the average network speed. aAverage doctor speed, where v is the doctor's response speed and T0 is the preset consultation duration;

[0135] Where k is a coefficient, obtained from the platform's preliminary auxiliary diagnosis, if (S y1 -S y2 ) / S y1 ≥50%; then k=1;

[0136] If (S) y1 -S y2 ) / S y1 <50%; and (S y2 -S y3 ) / S y3 ≥50%; then k=1+(S y1 -S y2 ) / S y1 ;

[0137] If (S) y1 -S y2 ) / S y1 <50%; and (S y2 -S y3 ) / S y3 <50%; then k = 1 + (S y1 +S y2 ) / (S y1 +S y2 +S y3 );

[0138] Offline appointment booking provides a demand selection window, which includes a first choice and a second choice;

[0139] The first option includes specialist appointments and regular medical visits;

[0140] The second option is doctor selection, including the doctor's gender and preference based on experience, attitude, or overall score;

[0141] If a patient chooses a regular appointment, the system automatically matches them with a doctor in the corresponding department and generates a queue number based on their second choice and the number of registered patients. The system updates the queue number information in real time and sends reminders. If a patient specifies a female doctor, but no female doctor is available for their chosen time slot, the system sends the patient the available time slots and / or hospitals in their selected area. Real-time updates include: for example, if a patient with an appointment on the 20th cancels their appointment 4 hours before the scheduled start time, the system promptly updates the queue number information for patients with appointments after the 20th and sends reminders. The system also promptly sends update reminders when there are changes in doctor schedules.

[0142] If a patient chooses to make an appointment with a specialist, a list of specialists, their introductions, and their schedules will be provided based on the hospital department the patient selects. The patient can then choose a suitable specialist and appointment time from the list.

[0143] The working principle of the above technical solution is as follows: It provides online and offline appointment options: patients can choose online consultation or offline appointment; if the patient chooses online consultation, the system will determine the consultation duration based on the patient's condition description and past medical records. The formula for calculating the consultation duration is:

[0144]

[0145] Where L is the network speed during consultation, L_a is the average network speed, n is the number of patients seen by the doctor at the same time, v_a is the average speed of the doctor, v is the response speed of the doctor, and T0 is the baseline consultation time.

[0146] The coefficient k is obtained based on the platform's preliminary auxiliary diagnosis; if (S y1 -S y2 ) / S y1 ≥50%; then k=1;

[0147] If (S) y1 -S y2 ) / S y1 <50%; and (S y2 -S y3 ) / S y3 ≥50%; then k=1+(S y1 -S y2 ) / S y1 ;

[0148] If (S) y1 -S y2 ) / S y1 <50%; and (S y2 -S y3 ) / S y3 <50%; then k = 1 + (S y1 +S y2 ) / (S y1 +S y2 +S y3 );

[0149] Offline appointments offer a selection window for your needs, including a first choice and a second choice. The first choice includes specialist appointments and regular consultations. The second choice is doctor selection, which can be based on the doctor's gender, experience, attitude, or overall rating.

[0150] If a patient chooses a regular appointment, the system will automatically match the corresponding department with a doctor based on the patient's second choice and the number of registered patients, and generate a queue number. The system will update the queue number in real time and send reminders to the patient. If the patient requests a female doctor, but there is no female doctor on duty for the selected time slot, the system will send the patient other time slots with female doctors on duty and / or hospital information. At the same time, the system will also send timely update reminders when there are changes in the doctor's schedule.

[0151] If a patient chooses to book an appointment with a specialist, the system will provide a list of specialists, their introductions, and their schedules based on the hospital and department the patient selects. The patient can then choose a suitable specialist and appointment time from the list.

[0152] The above technical solution offers the following benefits: Patients can choose their appointment method online or offline, avoiding the wasted time of traditional queuing. The system sets the consultation duration based on the patient's condition description and past medical records, enabling more accurate scheduling of doctors and improving consultation efficiency. Offline appointments offer multiple selection windows, allowing patients to choose between specialist appointments or regular appointments based on their needs, and to select a doctor based on gender, experience, attitude, or overall rating. This meets patients' personalized needs for appointment methods and doctor selection. For regular appointments, the system automatically matches doctors in the corresponding departments based on the patient's selection and the number of registered patients, generating queue information. The system updates queue information in real time and sends reminders to patients, including cancellations and changes in doctor schedules. Meanwhile, if a patient requests a female doctor's appointment time but no female doctor is available, other suitable time slots and / or hospitals are offered as options. For patients requiring specialist consultations, the system provides a list of specialists in the corresponding hospital departments, along with specialist introductions and scheduling information, allowing patients to choose a suitable specialist and appointment time based on their needs. By introducing intelligent algorithms to calculate consultation duration and coefficient k, the possibility of human error and oversight is reduced, improving the accuracy and reliability of the program. The formula considers multiple factors, such as network speed, the number and speed of doctors receiving consultations, etc., and through reasonable coefficients and calculation methods, the calculation of consultation duration is more scientific. This reduces unfairness caused by arbitrary scheduling of consultation durations, ensuring that patients receive fair consultation services. Network speed is the rate at which data is transmitted during the consultation. Considering that network speed can affect the smoothness and efficiency of online consultations, including the clarity of video calls and real-time audio transmission, network speed is included as a physical quantity, allowing for adjustment of consultation duration based on specific speed conditions. Average network speed is a reference value, representing the overall network quality of the platform. By comparing the current network speed with the average network speed, consultation duration can be further adjusted to adapt to different network environments. This ensures that patients receive a reasonable consultation duration under various network conditions. The number of patients a doctor sees simultaneously also affects consultation duration. An increase in the number of patients a doctor sees at the same time may lead to distraction or increased service pressure, thus prolonging consultation time. By considering the number of doctors seeing patients as a physical quantity, consultation duration can be adjusted according to specific circumstances to ensure patients receive adequate service. A doctor's working speed and response speed directly affect the efficiency and duration of consultations. By considering the average speed of doctors and the specific response speed of each doctor, consultation duration can be adjusted based on the individual doctor's ability and work efficiency to better meet patients' needs.

[0153] In summary, the appointment module of this smart healthcare online service system can improve the efficiency of medical treatment, meet personalized needs, update and remind patients in real time, reduce human error and omissions, and improve the convenience and accuracy of appointments.

[0154] This embodiment describes a smart healthcare online service system, wherein the patient visit statistics module includes:

[0155] Historical medical records statistics and appointment statistics;

[0156] Historical records include:

[0157] The number of patients visiting each department within each time period was counted by dividing the time period into time periods; online and offline visits were counted separately.

[0158] The day is divided into time periods by hour, and the number of patients in each department during each time period is counted.

[0159] The average number of patients visited by each department during each time period within a week is calculated; rest days and workdays are counted separately; Monday to Friday are workdays, and Saturday and Sunday are rest days;

[0160] By using the statistical data of each month over the past year as an observation window, the distribution of the number of patients in each department can be obtained, including peak time periods, off-peak time periods, and low time periods.

[0161] Based on the distribution, resources should be allocated reasonably, including arranging the working hours and number of doctors and nurses, considering increasing outpatient hours, arranging expert consultations, or opening evening outpatient clinics for specific departments.

[0162] The working principle of the above technical solution is as follows: Based on historical medical records, the number of patients in each department within each time period is statistically analyzed, categorized by time period and department. Online and offline medical visits are statistically analyzed separately to facilitate comparison and analysis of different treatment methods. Each day is divided into hourly intervals, and the number of patients in each department within each time period is statistically analyzed. This allows for a more detailed observation of the distribution of patients in each department and accurate identification of peak treatment times. The number of patients in each department within each time period within a week is statistically analyzed, separately for weekdays and rest days. This provides an understanding of the average number of patients in different time periods and on different types of days (weekdays or rest days). Data from the past year is used as an observation window to statistically analyze the number of patients in each department each month. Analysis reveals peak, off-peak, and low-peak hours for each department each month. This allows for informed resource allocation decisions, such as increasing the working hours and number of doctors and nurses, extending outpatient hours, or scheduling specialist consultations. Medical institutions gain a better understanding of patient distribution, including time periods, departments, and modes of access. Based on this data, resource allocation can be optimized to rationally arrange the working hours and number of doctors and nurses, providing more convenient and efficient medical services, such as extending outpatient hours or opening evening clinics for specific departments. This better meets patient needs and improves the service quality and efficiency of medical institutions.

[0163] The effects of the above technical solution are as follows: By dividing the data into time periods and departments for statistical analysis, medical institutions can more accurately understand the number of patients in each time period and department. This helps identify peak and relatively low-demand periods, providing a basis for the rational allocation of resources. Based on the distribution of patients, medical institutions can rationally arrange the working hours and numbers of doctors and nurses. For example, increasing medical staff during peak periods provides more adequate medical services; appropriately reducing staff during low-demand periods avoids resource idleness. By analyzing the average number of patients at different times of the week and on rest days / workdays, medical institutions can rationally arrange outpatient hours to meet patients' needs. In addition, based on the distribution of patients over the past year, medical institutions can also open evening clinics in specific departments or schedule specialist consultations, providing more flexible and efficient medical services. Rational allocation of resources and optimization of service arrangements can improve the service quality of medical institutions. Patients can receive timely medical care at appropriate times, reducing waiting time and improving the patient experience; at the same time, medical institutions can better meet patients' needs and improve the efficiency and level of medical services.

[0164] In conclusion, by analyzing historical medical records and the number of appointments, medical institutions can achieve precise resource allocation, improve service efficiency and quality, thereby better meeting patient needs and enhancing the overall operational effectiveness of medical institutions.

[0165] This embodiment discloses a smart healthcare online service system, wherein the appointment service module includes:

[0166] Schedule VR online guidance services and in-home services;

[0167] For patients with limited mobility or elderly patients, such as those needing dressing changes for external injuries, they can make an appointment with a community doctor to provide home service; they can also make an appointment for online service and receive VR online guidance for dressing changes.

[0168] For home visits, the appointment service module obtains the patient's address or location information and the services requested. Based on the address or location information and the services requested by the patient, the system recommends the nearest community doctor to schedule a home visit.

[0169] The working principle and effect of the above technical solution are as follows: Patients can select VR online guidance services through the appointment service module. First, the patient provides relevant personal information and a description of their condition. The system will match a suitable doctor or specialist based on the patient's location or the settings of the medical institution. The patient can choose the appointment date and time slot and pay the corresponding fee. During the appointment time slot, the patient can receive online guidance services through VR devices. The doctor or specialist will provide detailed guidance and suggestions. For patients with limited mobility or elderly patients requiring wound dressing changes, they can choose to schedule home visits. The patient provides their address or location information and a description of the service in the appointment service module. The system will recommend nearby community doctors to provide home visits based on the patient's location or the settings of the medical institution. The patient can choose the appointment date and time slot and pay the corresponding fee. After accepting the appointment, the community doctor will go to the patient's home at the appointed time and location to provide the corresponding medical services.

[0170] The core principle of the appointment service module is to match suitable doctors or specialists based on the information provided by the patient, and arrange appointment times and service methods according to the patient's choices and payment arrangements. For VR online guidance services, virtual reality technology is used to enable remote communication and guidance between doctors and patients. For home visits, nearby community doctors are matched, and doctors are arranged to provide home visits according to the appointment time and location. This appointment service module can provide more convenient and flexible medical services, meeting the needs of patients with limited mobility or those requiring remote guidance, while improving the efficiency of medical resource utilization.

[0171] This embodiment describes a smart healthcare online service system, wherein the evaluation module includes:

[0172] System evaluation feedback and physician evaluation;

[0173] The system evaluation feedback includes evaluation of system response speed, usability, error feedback, and improvement suggestions;

[0174] The doctor evaluation includes evaluations of online consultations and offline visits, and the evaluation includes evaluations of professional competence, communication skills, service attitude, and response speed.

[0175] All patients can provide feedback; feedback can be given online to the doctors who provided in-person care; feedback can be given within one week after the consultation.

[0176] The working principle and effect of the above technical solution is to allow users to evaluate the system and doctors in order to collect user feedback and opinions.

[0177] System evaluation feedback:

[0178] a. System response speed evaluation: Users can evaluate the system's response speed, such as rapid response and appropriate waiting time.

[0179] b. Usability Evaluation: Users can evaluate the system's ease of use and user experience, including aspects such as interface design and function settings.

[0180] c. Error Feedback: If users encounter errors or problems while using the system, they can provide error feedback to help improve the system's stability and functionality.

[0181] d. Improvement suggestions: Users can provide suggestions for improving the system, including adding new features and optimizing the user experience.

[0182] Doctor's evaluation:

[0183] a. Online Consultation Doctor Evaluation: Patients can evaluate their online consultation doctors. Evaluations can include aspects such as the doctor's professional competence, communication skills, service attitude, and response speed.

[0184] b. Offline consultation evaluation: Patients can evaluate the doctors they booked online. Evaluations can include the doctor's professional competence, communication skills, service attitude, and response speed.

[0185] Evaluation subjects and time:

[0186] a. All users can rate the system and doctors, whether they consult online or in person.

[0187] b. The evaluation period is within one week after the consultation, giving users sufficient time to provide feedback.

[0188] These evaluations can help improve system performance and user experience, and provide a reference for the quality of doctor services. Users can rate the system based on their actual experience, including its responsiveness, ease of use, error feedback, and suggestions for improvement. They can also evaluate the doctor's professional competence, communication skills, service attitude, and response speed. This allows for continuous improvement of the system and the quality of doctor services, providing better smart healthcare online services.

[0189] This embodiment describes a smart healthcare online service system, wherein the optimization module includes:

[0190] System function optimization, service optimization, and model optimization;

[0191] The service optimization includes personnel training;

[0192] The model optimization includes:

[0193] The accuracy of model-assisted diagnosis is obtained by comparing the doctor's diagnosis results with the auxiliary diagnosis suggestions provided by the platform.

[0194]

[0195] Where k is a coefficient, obtained from the platform's preliminary auxiliary diagnosis, F Mj For primary diagnostic similarity; F Cj For secondary diagnostic similarity, F Zj The similarity of treatment recommendations is represented by m, which is the number of cases. The update cycle can be one day, where m is the original number of cases plus the number of newly added cases with diagnostic results.

[0196] Compare the primary diagnoses: Check whether the primary diagnoses are consistent or differing.

[0197] Compare secondary diagnoses: If there are doctors and platforms that offer suggestions for secondary diagnoses, you can also make comparisons;

[0198] Compare treatment recommendations: Compare the doctor's treatment recommendations with the platform's auxiliary diagnostic recommendations to see if there are any differences in medication recommendations, surgical recommendations, or other treatment measures;

[0199] Optimize the model regularly.

[0200] The working principle of the above technical solution is as follows: Regularly evaluate and improve system functions, optimizing function settings and interface design based on user feedback and needs; promptly fix and improve bugs and errors in the system to ensure system stability and reliability. Conduct personnel training to improve doctors' professional capabilities and service quality; training content includes communication skills, diagnostic abilities, and service attitude to provide a better medical service experience; compare doctors' diagnostic results with the auxiliary diagnostic suggestions provided by the platform to obtain the accuracy rate of model-assisted diagnosis; calculate a weighted sum based on the similarity of primary diagnoses, secondary diagnoses, and treatment suggestions, where the coefficient k is obtained from the platform's initial auxiliary diagnosis, the update cycle can be set to one day, determined by the number of consultations, and m represents the number of cases in the original model plus the number of newly added cases with diagnostic results; regularly use this case data to optimize and adjust the model, improving the accuracy and reliability of auxiliary diagnosis.

[0201] Based on the above working principles, the system can continuously optimize its functions, services, and models, providing a better user experience, improving the quality of doctors' services, and continuously improving the accuracy of the model's assisted diagnosis, thereby providing accurate, efficient, and convenient smart healthcare online services.

[0202] The effects of the above technical solution are as follows: Through personnel training, doctors' professional abilities and service attitudes are improved, enabling them to provide users with more accurate, timely, and user-friendly medical consultations and auxiliary diagnostic services, thus enhancing user satisfaction. By comparing doctors' diagnostic results with the auxiliary diagnostic suggestions provided by the platform, and calculating the weighted sum of the similarity of primary diagnoses, secondary diagnoses, and treatment suggestions, the accuracy of the model's auxiliary diagnosis is obtained. This allows for the evaluation and optimization of model accuracy and the provision of more reliable auxiliary diagnostic results. Based on the update cycle (which can be set to one day), newly added cases with diagnostic results are added to the original model's case count, achieving continuous optimization and updates to the model. This allows for the continuous accumulation of more case data, improving the model's accuracy and generalization ability. By setting the coefficient k, the similarity of primary diagnoses, secondary diagnoses, and treatment suggestions is comprehensively weighed based on the platform's preliminary auxiliary diagnostic results. In the formula, m represents the number of cases, and the update cycle can be set to one day. This allows the number of newly added cases with diagnostic results to be added to the original model's case count, enabling the model to obtain more training data and feedback information in a timely manner. By continuously accumulating case data, the accuracy and generalization ability of the model can be improved. This allows for adjustments to the importance of various indicators based on specific needs and initial platform assessments, making the model more aligned with real-world application scenarios. The optimized smart healthcare online service system can provide more accurate and reliable auxiliary diagnostic results, helping doctors and users better understand their conditions and offering more appropriate treatment suggestions. Simultaneously, the system can be continuously updated and optimized to adapt to evolving medical knowledge and case data, providing more efficient and convenient online services.

[0203] This embodiment provides a smart healthcare online service method, the method comprising:

[0204] S1. Obtain the patient's medical condition description and past medical records; automatically generate the patient's health record based on the medical condition description and past medical records; and provide self-service, intelligently providing preliminary auxiliary diagnostic suggestions;

[0205] S2. Provide online appointment for medical treatment through online medical services, and provide offline appointment for medical treatment through offline appointment services;

[0206] S3. The system uses the medical visit statistics module to collect historical medical visit records and the number of appointments. Based on the results of the historical medical visit records, resources are allocated reasonably. Based on the results of the appointments, user needs, and doctor scheduling, doctors are automatically matched.

[0207] S4. Provide appointment-based VR online guidance and on-site services through the appointment service module;

[0208] S5. Obtain system evaluation feedback, doctor evaluation, and home service evaluation through the evaluation module;

[0209] S6. Optimize system functions, services, and models through the optimization module.

[0210] The working principle of the above technical solution is as follows: Acquiring patient information: First, the system obtains the patient's description of their condition and past medical records. Based on this information, the system automatically generates the patient's health record and performs a preliminary auxiliary diagnosis. This step helps the user prepare the necessary documents for the consultation and provides a foundation for subsequent diagnosis and treatment. The system meets the user's medical needs through online and offline appointment services. Users can choose to make an appointment online or offline for face-to-face treatment at a medical institution. The system uses a consultation statistics module to statistically analyze historical consultation records and count the number of appointments. Based on the statistical results, the system rationally allocates resources, including doctor scheduling and the allocation of medical equipment. The system automatically matches suitable doctors based on the number of appointments and user needs, ensuring effective resource utilization. The appointment service module meets different user needs by providing VR online guidance and home visit services. Users can choose to receive remote guidance and treatment suggestions from doctors through virtual reality technology; or they can choose home visit services where doctors personally visit their homes for treatment. The system obtains user feedback on the system, doctors, and home visit services through an evaluation module. These evaluations and feedback help understand user satisfaction with the service and identify areas for improvement, providing a reference for subsequent system optimization. The optimization module refines system functions, service processes, and models based on the collected feedback. Through continuous optimization, system performance and user experience are improved, as well as the accuracy and effectiveness of assisted diagnosis. In summary, this smart healthcare online service method integrates information and resources, provides convenient appointment services and assisted diagnostic suggestions, meets users' medical needs, and continuously improves system performance and user experience through optimization.

[0211] The effects of the above technical solution are as follows: Users can choose to make appointments online or offline, eliminating the need for traditional queuing. This significantly reduces patient waiting time and improves efficiency. The system generates health records based on the patient's condition description and past medical records, and automatically provides preliminary auxiliary diagnostic suggestions. This helps users obtain reference opinions before making an appointment and better understand their condition. By statistically analyzing historical medical records and the number of appointments, the system can rationally allocate doctor and medical equipment resources. This ensures effective resource utilization and improves the accessibility of medical services. The system automatically matches suitable doctors based on the number of appointments and user needs. Furthermore, the appointment service module provides online VR guidance and home visit services, offering users a more convenient way to seek medical care. The evaluation module obtains user feedback on the system, doctors, and home visit services. This helps understand user satisfaction and areas for improvement, optimize system functions and service processes, and enhance user experience. The system optimizes its functions, services, and models through the optimization module and continuously improves based on user feedback. Continuous optimization can improve system performance and user experience, providing more accurate and efficient auxiliary diagnostic services. This smart healthcare online service approach can provide convenient access to medical care, auxiliary diagnostic suggestions, rational allocation of resources, doctor matching and guidance services, and improve system performance and user experience through evaluation feedback and continuous optimization.

[0212] This embodiment provides a smart healthcare online service method, wherein step S1 includes:

[0213] The collected descriptions of the patient's condition and past medical records are stored in a structured format in the database;

[0214] Create corresponding tables in the database, including a patient information table, a disease description table, and a treatment record table, and link the tables through primary key-foreign key relationships;

[0215] Data preprocessing of disease descriptions and past medical records yields preprocessed information, including text cleaning and standardization;

[0216] Natural Language Processing (NLP) technology is used to transform preprocessed text information into vector representations and extract useful features; the useful features include symptoms, keywords, and medical history information.

[0217] Extract numerical data from medical records;

[0218] Cluster analysis was performed on a large number of cases on the platform, and combined with doctors' diagnostic results, to obtain a set of cluster features for each disease;

[0219] The newly acquired disease features are compared with the features in the cluster feature set to obtain the similarity S. yi;

[0220]

[0221] Where w1, w2...wn are the weights of the features, S1, S2...Sn are the similarities of each feature, and n is the number of features included; n = min(P, Q), where P is the newly acquired disease feature, and Q is the number of features corresponding to a certain disease in the feature library;

[0222] For different disease feature databases, different similarity scores are obtained, and the top three similarity scores S are... y1 S y2 S y3 ;

[0223] If (S) y1 -S y2 ) / S y1 ≥50%; then S y1 The corresponding disease is given suggestions as a preliminary auxiliary diagnosis;

[0224] If (S) y1 -S y2 ) / S y1 <50%; and (S y2 -S y3 ) / S y3 ≥50%; then S y1 and S y2 The corresponding diseases are given suggestions as preliminary auxiliary diagnoses; and the probabilities are compared based on the degree of similarity.

[0225] If (S) y1 -S y2 ) / S y1 <50%; and (S y2 -S y3 ) / S y3 <50%; then S y1 S y2 S y3 The corresponding disease is given as a preliminary auxiliary diagnosis, and the probability comparison results are given based on the similarity.

[0226] The working principle of the above technical solution is as follows: Acquire the patient's medical condition description and past medical records, including voice descriptions, text descriptions, photos, and / or examination reports; store the collected medical condition descriptions and past medical records in a structured form in a database, establishing corresponding patient information tables, medical condition description tables, and medical record tables, and linking them through primary key-foreign key relationships; preprocess the data in the medical condition description and past medical record tables, including text cleaning and standardization, to ensure data quality and consistency; use Natural Language Processing (NLP) technology to convert the preprocessed text information into vector representations and extract useful features, such as symptoms, keywords, and medical history information. Feature extraction can be performed using methods such as the bag-of-words model and Word2Vec. Numerical data from medical records is extracted, and time-series data is analyzed using time-series analysis methods to extract corresponding features, such as statistical features, frequency domain features, and time domain features. Cluster analysis is performed on a large number of case features on the platform, combined with doctors' diagnostic results, to obtain a cluster feature set for each disease. The newly acquired disease features are compared with the features in the cluster feature set to calculate similarity. Based on the similarity, preliminary auxiliary diagnostic suggestions are given. Depending on the situation, one disease may be suggested, or multiple diseases may be suggested, with a comparison of probabilities based on similarity. Feature weights can be determined based on clinical experience, statistical analysis, machine learning methods, and expert consensus to ensure the accuracy and reliability of the auxiliary diagnosis.

[0227] The effects of the above technical solution are as follows: By combining the patient's description of their condition and past medical records with the platform's intelligent preliminary auxiliary diagnostic suggestions, the accuracy of diagnosis can be improved; the system uses natural language processing technology to convert text information into vector representations and extract useful features, helping doctors better understand the patient's symptoms and disease progression; patients can conduct self-diagnosis anytime, anywhere through the smart healthcare platform, without waiting for appointments or queuing; it reduces resource waste for some mild cases, allowing patients to describe their condition in detail through voice, text, and uploaded photos, providing doctors with more comprehensive information and thus improving diagnostic efficiency; based on the patient's description of their condition and past medical records, the system will provide preliminary auxiliary diagnostic suggestions based on cluster analysis and similarity calculations; based on the similarity score, the system will provide a probability comparison result according to certain rules, helping patients understand various possible diseases and make appropriate choices; the system stores the collected descriptions of conditions and past medical records in a structured form in a database and manages them through a database management system. This facilitates subsequent data analysis and research, helps discover potential disease patterns, and provides more accurate diagnostic suggestions. The weight parameters in the formula can be adjusted based on clinical experience, statistical analysis, machine learning methods, and expert consensus to reflect the importance of different features for diagnosis. By setting appropriate weights, the similarity calculation can be made more accurate and reliable. The weight parameters in the formula can be adjusted according to the importance of different features, thus taking multiple feature factors into account and improving the comprehensiveness and accuracy of the diagnosis. Similarity is an indicator that measures the degree of similarity between newly acquired disease features and a set of features from a certain disease cluster. By calculating similarity, the degree of matching between new features and existing disease features can be assessed.

[0228] P represents the number of newly acquired disease features, and Q represents the number of features in the feature set for a certain disease. By comparing the magnitudes of P and Q, the number of features used for similarity calculation is determined to ensure an appropriate number, avoiding the influence of too many or too few features on the results. The feature weights (w1, w2, ..., wn) represent the importance of each feature in the similarity calculation. Weight values ​​can be determined by combining clinical experience, statistical analysis, machine learning methods, and expert consensus. A higher weight means that the feature has a greater influence in the similarity calculation. Based on the magnitude of similarity, the diseases are ranked, and the three most similar diseases are selected as candidates. These similarities can serve as preliminary auxiliary diagnostic suggestions. If the first-ranked similarity is significantly greater than the second-ranked similarity, it indicates a high correlation with the corresponding disease, while other similarities have low correlation; in this case, the first-ranked similarity is used as the preliminary auxiliary diagnostic suggestion. If the first and second-ranked similarities are not significantly different, both possibilities are relatively high; therefore, diseases corresponding to both similarities are used as preliminary auxiliary diagnostic suggestions, ranked by similarity. If the first, second, and third-ranked similarities are not significantly different, diseases corresponding to all three similarities are used as preliminary auxiliary diagnostic suggestions, ranked by similarity. Based on the above formula, similarity can be calculated between newly acquired disease features and existing cluster feature sets. Preliminary auxiliary diagnostic suggestions can be given based on the magnitude of the similarity and the comparison of differences. This helps doctors identify possible diseases more quickly and accurately, and provide corresponding diagnostic suggestions and probability comparison results.

[0229] In summary, the self-service module of the smart healthcare online service system can provide fast, accurate, and personalized diagnostic suggestions, allowing patients to access medical services anytime, anywhere, and improving the efficiency and accuracy of doctors' diagnoses. Furthermore, through data management and analysis, it can also provide valuable data resources for medical research.

[0230] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A smart healthcare online service system, characterized in that, The system includes: Service platforms, terminal equipment, community service stations, and hospital service terminals; The service platform connects community service stations and hospital service terminals; The terminal device is used for user registration and access to the service platform; The service platform includes a registration and login module, a self-service module, an appointment module, an appointment statistics module, an appointment service module, an evaluation module, and an optimization module.

2. The smart medical online service system according to claim 1, characterized in that, The registration and login module includes: Users register account information through the registration module. After registration, they can select to add patient information, which includes name, gender, age, contact person, address, ID number, and emergency contact person. Based on the patient's ID number, the patient's social security account is directly linked. Establish health records for patients.

3. The smart medical online service system according to claim 1, characterized in that, The self-service module includes the following methods: Obtain the patient's medical history and past medical records; Based on the description of the illness and past medical records, the platform intelligently provides preliminary auxiliary diagnostic suggestions; Based on the aforementioned auxiliary suggestions, patients may choose to purchase medication directly, consult a doctor, or undergo observation; the description of the condition includes voice description, text description, photos, and / or examination reports; Based on the description of the illness and past medical records, the platform intelligently provides preliminary auxiliary diagnostic suggestions; The collected descriptions of the patient's condition and past medical records are stored in a structured format in the database; Create corresponding tables in the database, including a patient information table, a disease description table, and a treatment record table, and link the tables through primary key-foreign key relationships; Data preprocessing of disease descriptions and past medical records yields preprocessed information, including text cleaning and standardization; Natural Language Processing (NLP) technology is used to transform preprocessed text information into vector representations and extract useful features; the useful features include symptoms, keywords, and medical history information. Extract numerical data from medical records; Cluster analysis was performed on a large number of cases on the platform, and combined with doctors' diagnostic results, to obtain a set of cluster features for each disease; The newly acquired disease features are compared with the features in the cluster feature set to obtain the similarity S. yi ; Where w1, w2...wn are the weights of the features, S1, S2...Sn are the similarities of each feature, and n is the number of features included; n = min(P, Q), where P is the newly acquired disease feature, and Q is the number of features corresponding to a certain disease in the feature library; For different disease feature databases, different similarity scores are obtained, and the top three similarity scores S are... y1 S y2 S y3 ; if( y1 -S y2 ) / S y1 ≥50%; then S y1 The corresponding disease is given suggestions as a preliminary auxiliary diagnosis; if( y1 -S y2 ) / S y1 <50%; and ( y2 -S y3 ) / S y3 ≥50%; then S y1 and S y2 The corresponding diseases are given suggestions as preliminary auxiliary diagnoses; and the probabilities are compared based on the degree of similarity. If (S) y1 -S y2 ) / S y1 <50%; and (S y2 -S y3 ) / S y3 <50%; then S y1 S y2 S y3 The corresponding disease is given as a preliminary auxiliary diagnosis, and the probability comparison results are given based on the similarity.

4. The smart medical online service system according to claim 3, characterized in that, The appointment booking module includes: Offers the option of online or offline appointment booking; If a patient chooses online consultation, the system sets the consultation duration based on the patient's description of their condition and past medical records; the consultation duration is: Where L represents the network speed during the consultation, L a Let n be the average network speed, n be the number of patients a doctor sees simultaneously, and v be the average network speed. a Average doctor speed, v is the doctor's response speed, and T0 is the preset time; Where k is a coefficient, obtained from the platform's preliminary auxiliary diagnosis, if (S y1 -S y2 ) / S y1 ≥50%; then k=1; If (S) y1 -S y2 ) / S y1 <50%; and (S y2 -S y3 ) / S y3 ≥50%; then k=1+(S y1 -S y2 ) / S y1 ; If (S) y1 -S y2 ) / S y1 <50%; and (S y2 -S y3 ) / S y3 <50%; then k = 1 + (S y1 +S y2 ) / (S y1 +S y2 +S y3 ); Offline appointment booking provides a demand selection window, which includes a first choice and a second choice; The first option includes specialist appointments and regular medical visits; The second option is the selection of a doctor, including the doctor's gender and preference based on experience, attitude, or overall score; If a patient chooses a regular appointment, the system will automatically match the corresponding department and doctor based on the patient's second choice and the number of registered patients, and generate a queue number; the queue number information will be updated in real time and a reminder will be sent. If a patient chooses to make an appointment with a specialist, a list of specialists, their introductions, and their schedules will be provided based on the hospital department the patient selects. The patient can then choose a suitable specialist and appointment time from the list.

5. The smart medical online service system according to claim 1, characterized in that, The medical visit statistics module includes: Historical medical records statistics and appointment statistics; Historical records include: The number of patients visiting each department within each time period was counted by dividing the time period into time periods; online and offline visits were counted separately. The day is divided into time periods by hour, and the number of patients in each department during each time period is counted. The average number of patients visited by each department during each time period within a week is calculated; statistics are compiled separately for rest days and workdays; By using the statistical data of each month over the past year as an observation window, the distribution of the number of patients in each department can be obtained, including peak time periods, off-peak time periods, and low time periods. Based on the distribution, resources should be allocated reasonably, including arranging the working hours and number of doctors and nurses, considering increasing outpatient hours, arranging expert consultations, or opening evening outpatient clinics for specific departments.

6. The smart medical online service system according to claim 1, characterized in that, The appointment service module includes: Schedule VR online guidance services and in-home services; For in-home services, the appointment service module obtains the patient's address or location information and the services requested. Based on the address or location information provided by the patient and the services requested, it displays information about doctors or nurses at nearby community hospitals, including their names, titles, and areas of expertise. The patient then makes an appointment based on this information.

7. The smart medical online service system according to claim 1, characterized in that, The evaluation module includes: System evaluation feedback, doctor evaluation, and home visit service evaluation; The system evaluation feedback includes evaluation of system response speed, usability, error feedback, and improvement suggestions; The doctor evaluation includes evaluations of online consultations and offline visits, and the evaluation includes evaluations of professional competence, communication skills, service attitude, and response speed.

8. The smart medical online service system according to claim 4, characterized in that, The optimization module includes: System function optimization, service optimization, and model optimization; The service optimization includes personnel training; The model optimization includes: The accuracy of model-assisted diagnosis is obtained by comparing the doctor's diagnosis results with the auxiliary diagnosis suggestions provided by the platform. Where k is a coefficient, obtained from the platform's preliminary auxiliary diagnosis, F Mj For primary diagnostic similarity; F Cj For secondary diagnostic similarity, F Zj The similarity of treatment recommendations is represented by m, which is the number of cases. The update cycle can be one day, where m is the original number of cases plus the number of newly added cases with diagnostic results. Compare the primary diagnoses: Check whether the primary diagnoses are consistent or differing. Compare secondary diagnoses: If there are doctors and platforms that offer suggestions for secondary diagnoses, you can also make comparisons; Compare treatment recommendations: Compare the doctor's treatment recommendations with the platform's auxiliary diagnostic recommendations to see if there are any differences in medication recommendations, surgical recommendations, or other treatment measures; Optimize the model regularly.

9. The method for providing smart healthcare online services according to claim 1, characterized in that, The method includes: S1. Obtain the patient's medical condition description and past medical records; automatically generate the patient's health record based on the medical condition description and past medical records; and provide self-service, intelligently providing preliminary auxiliary diagnostic suggestions; S2. Provide online appointment for medical treatment through online medical services, and provide offline appointment for medical treatment through offline appointment services; S3. The system uses the medical visit statistics module to collect historical medical visit records and the number of appointments. Based on the results of the historical medical visit records, resources are allocated reasonably. Based on the results of the appointments, user needs, and doctor scheduling, doctors are automatically matched. S4. Provide appointment-based VR online guidance and on-site services through the appointment service module; S5. Obtain system evaluation feedback, doctor evaluation, and home service evaluation through the evaluation module; S6. Optimize system functions, services, and models through the optimization module.

10. A smart healthcare online service method according to claim 9, characterized in that, S1 includes: The collected descriptions of the patient's condition and past medical records are stored in a structured format in the database; Create corresponding tables in the database, including a patient information table, a disease description table, and a treatment record table, and link the tables through primary key-foreign key relationships; Data preprocessing of disease descriptions and past medical records yields preprocessed information, including text cleaning and standardization; Natural Language Processing (NLP) technology is used to transform preprocessed text information into vector representations and extract useful features; the useful features include symptoms, keywords, and medical history information. Extract numerical data from medical records; Cluster analysis was performed on a large number of cases on the platform, and combined with doctors' diagnostic results, to obtain a set of cluster features for each disease; The newly acquired disease features are compared with the features in the cluster feature set to obtain the similarity S. yi ; Where w1, w2…wn are the weights of the features, S1, S2…Sn are the similarities of each feature, and n is the number of features included; n = min(P, Q), where P is the newly acquired disease feature, and Q is the number of features corresponding to a certain disease in the feature library; For different disease feature databases, different similarity scores are obtained, and the top three similarity scores S are... y1 S y2 S y3 ; if( y1 -S y2 ) / S y1 ≥50%; then S y1 The corresponding disease is given suggestions as a preliminary auxiliary diagnosis; If (S) y1 -S y2 ) / S y1 <50%; and (S y2 -S y3 ) / S y3 ≥50%; then S y1 and S y2 The corresponding diseases are given suggestions as preliminary auxiliary diagnoses; and the probabilities are compared based on the degree of similarity. If (S) y1 -S y2 ) / S y1 <50%; and (S y2 -S y3 ) / S y3 <50%; then S y1 S y2 S y3 The corresponding disease is given as a preliminary auxiliary diagnosis, and the probability comparison results are given based on the similarity.

Images