Artificial intelligence hospital outpatient service self-service system data access device and method

Abstract

The invention provides an artificial intelligence hospital outpatient service self-service system data access device and method, and relates to the technical field of hospital outpatient service data access. The method comprises the steps of receiving an outpatient service obtaining request sent by a patient user, analyzing and querying whether an authorization token exists, and authorizing the service according to the token. The authorization token evaluation predicts the daily outpatient quantity of the department by establishing an artificial intelligence outpatient quantity model and sets the daily outpatient quantity as a threshold value; and online predicting the doctor seeing time of the patient by establishing an artificial intelligence department waiting time model. And if the doctor seeing time of the patient is one hour before the end time of the workday of the current department, the threshold value is automatically changed into the actual triage amount of the current day. According to the method, the outpatient amount can be effectively adjusted at the client, unnecessary load of a hospital background system is avoided, and common outpatient appointment and emergency appointment are uniformly adjusted and controlled at the client for the first time, so that the outpatient service level of a hospital is improved, ordered triage treatment is realized, and blind waiting time of patients and workload of doctors are reduced.

Description

Technical Field

[0001] The present invention relates to the field of hospital outpatient data access technology, and in particular to an artificial intelligence hospital outpatient self-service system data access device and method. Background Art

[0002] Currently, when patients use mobile apps or self-service terminal data collection devices in hospitals to make appointments, register, and other outpatient activities, after taking a number, they need to go to the designated department to swipe the number and wait for the doctor's outpatient clinic. Since the number of patients waiting and the waiting time in the triage department of the hospital are very uncertain, patients often wait in disorder. Especially in large hospitals, such as tertiary hospitals, after patients make an appointment and take a number in the morning, they often have to wait for a whole morning or even a day of work to complete the medical activities. In addition, for hospitals, the uncertainty and uncontrollability of the number of patients visiting the triage department every day make it impossible to accurately predict the number of medical staff on duty in the department, etc. These situations seriously affect the hospital's service efficiency and patient satisfaction with medical treatment. Therefore, on the one hand, patients often have to wait for 5-7 hours for a 15-minute consultation, and on the other hand, doctors face overcrowded patients and work for long hours under tremendous pressure.

[0003] The existing self-service terminal appointment payment and settlement system and department triage queuing system only passively respond to blind and unmonitored patient flow data, and fail to truly solve the problem of long waiting times for patients in hospitals.

[0004] Based on this, this application is filed. Summary of the invention

[0005] The purpose of the present invention is to provide a data access method and device based on an artificial intelligence hospital outpatient self-service terminal, so as to realize orderly hospital treatment, reduce the blind waiting time of patients, reduce the workload of hospital doctors, and improve hospital service efficiency.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] According to one aspect of the present invention, a method for accessing data of an artificial intelligence hospital outpatient self-service system is provided, the method comprising: receiving a request for obtaining a service sent by a terminal user; parsing the request for obtaining a service, querying whether the user has an authorization token, and authorizing or not authorizing the user to obtain the service according to the token status; when the user does not have an authorization token, querying the corresponding database according to the content of the service request and performing evaluation and analysis to obtain an evaluation result; giving an authorization or non-authorization token according to the evaluation and analysis result and sending it to the terminal; wherein the service at least includes a unified appointment service, and when receiving a request for obtaining an appointment service sent by a terminal user, parsing the request for obtaining a service, querying whether the user has an authorization token, if there is no authorization token, obtaining the corresponding department outpatient clinic according to the content of the unified appointment service request, querying the outpatient volume threshold and the outpatient volume scheduled for the day of the department outpatient clinic, performing evaluation and analysis according to the outpatient volume threshold and the outpatient volume scheduled for the day, obtaining an authorization or non-authorization token and sending it to the terminal; the outpatient volume for the day is obtained by establishing a department outpatient volume prediction model.

[0008] Another aspect of the present invention provides a data access device for an artificial intelligence hospital outpatient self-service system, the device comprising: a data receiving module for receiving a service acquisition request sent by a terminal user; a request authentication module for parsing the service acquisition request, querying whether the user has an authorization token, and authorizing or not authorizing the user to obtain the service according to the token situation; an evaluation and analysis module for querying the corresponding database and performing evaluation and analysis according to the content of the service request when the user does not have an authorization token, and obtaining an evaluation result; an evaluation result sending module for giving an authorization or non-authorization token according to the evaluation and analysis result and sending it to the terminal; wherein the service at least includes a unified appointment service, and when receiving the appointment service acquisition request sent by the terminal user, the service acquisition request is parsed, the user is queried for an authorization token, if there is no authorization token, the corresponding department outpatient clinic is acquired according to the content of the unified appointment service request, the outpatient volume threshold and the appointment outpatient volume of the department outpatient clinic on the same day are queried, and the evaluation and analysis are performed according to the outpatient volume threshold and the appointment outpatient volume on the same day, and the authorization or non-authorization token is obtained and sent to the terminal; the outpatient volume on the same day is obtained by establishing a department outpatient volume prediction model.

[0009] The above-mentioned artificial intelligence hospital outpatient self-service system data access method and device can effectively control the number of appointments in hospital departments by predicting the outpatient volume of the day and setting the outpatient volume threshold of the day, thereby achieving orderly hospital visits, reducing patients' blind waiting time, reducing the workload of hospital doctors, and improving hospital service efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0011] Figure 1 A flowchart of a data access method for an artificial intelligence hospital outpatient self-service system provided in a specific embodiment of the present invention;

[0012] Figure 2 A schematic diagram of the functional modules of a data access device for an artificial intelligence hospital outpatient self-service system provided in a specific embodiment of the present invention;

[0013] Figure 3 A schematic diagram of data flow (microservice authentication architecture) in a data access method for an artificial intelligence hospital outpatient self-service system provided in a specific embodiment of the present invention;

[0014] Figure 4 This is an information interaction diagram of each microservice and each database, intelligent gateway, and processing center in the data access device of the artificial intelligence hospital outpatient self-service system provided by a specific embodiment of the present invention;

[0015] Figure 5 This is an algorithm flow chart of the analysis module in the data access device of the artificial intelligence hospital outpatient self-service system provided by a specific implementation method of the present invention.

[0016] Figure numerals: collection terminal 100, data collection module 110, request sending module 120, intelligent gateway 200, data receiving module 210, request authentication module 220, information feedback module 230, processing center 300, evaluation receiving module 310, analysis module 320, evaluation result sending module 330. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. This article will take the microservice authentication framework as an example to specifically describe the solution of the present invention, but it is not limited to this. The solution of the present invention can also be applied to distributed application systems, serverless cloud systems, etc.

[0018] Please refer to Figure 1, a flow chart of the data access method of the artificial intelligence hospital outpatient self-service system provided by the present invention.

[0019] Step S100: receiving a microservice acquisition request sent by a terminal user.

[0020] Step S200: parse the request for obtaining the microservice, query whether the user has an authorization token, and authorize or not authorize the user to obtain the microservice according to the token status.

[0021] Step S300, when the user does not have an authorization token, query the corresponding database according to the content of the microservice request and perform evaluation analysis to obtain an evaluation result;

[0022] Step S400, giving an authorization or non-authorization token according to the evaluation and analysis result and sending it to the terminal;

[0023] Among them, the microservice at least includes a unified appointment microservice. When receiving a request for obtaining an appointment microservice sent by a terminal user, the microservice request is parsed and the user is queried whether there is an authorization token. If there is no authorization token, the corresponding department outpatient clinic is obtained according to the content of the unified appointment microservice request, and the daily outpatient volume threshold and the daily scheduled outpatient volume of the department outpatient clinic (which have already occurred at the time of the request) are queried. Based on the daily outpatient volume threshold and the daily scheduled outpatient volume, an evaluation and analysis is performed, and an authorization or non-authorization token is obtained and sent to the terminal; the daily outpatient volume is obtained by establishing a department outpatient volume prediction model.

[0024] In one embodiment, please refer to Figure 2 , the data access method of the artificial intelligence hospital outpatient self-service system provided by the present invention is described by taking the acquisition terminal 100, the intelligent gateway 200, and the processing center 300 as examples. It can be understood by ordinary technicians in this field that the described technical solution is also applicable to other intelligent devices. The acquisition terminal 100 mainly includes a data acquisition module 110 and a request sending module 120; the intelligent gateway 200 mainly includes a data receiving module 210, a request authentication module 220, and an information reverse module 230; the processing center 300 includes an evaluation receiving module 310, an analysis module 320, and an evaluation result sending module 330.

[0025] The patient logs into the system through the collection terminal 100, i.e., the client (such as self-service machine A / self-service machine C or mobile terminal B or mobile terminal D) by using an ID card, medical card, or face recognition. The patient first sends a micro-service request (such as a same-day appointment for a department outpatient clinic). After receiving the request, the intelligent gateway 200 identifies the requested micro-service content. The intelligent gateway 200 sends a processing information request to the processing center 300. After receiving the processing information request, the processing center 300 contacts the micro-service and the relevant database through the network routing to obtain the department outpatient history data (see Figure 4 ).

[0026] In one embodiment, the processing center 200 establishes a time series prediction model (department outpatient / triage volume prediction model) to predict the outpatient volume of the department on that day as the outpatient volume threshold of the day (the first threshold of the outpatient volume on that day). The department outpatient volume prediction model adopts an LSTM neural network algorithm model, inputs the historical outpatient data of the department's patients in the past year, and predicts the department's triage outpatient volume for one day or one week. If the department's scheduled outpatient volume on that day is lower than the first threshold of the outpatient volume on that day, the information of the token permission instruction is fed back to the intelligent gateway 200, so that the intelligent gateway 200 sends an authorization token (token) to the terminal; if the department's scheduled outpatient volume on that day exceeds the threshold, the token authorization instruction is fed back to the token authorization authentication center (authentication request module 220) of the intelligent gateway 200, so that the intelligent gateway 200 sends a non-authorization token to the terminal, and a notification message that the department's outpatient clinic on that day has been full.

[0027] In one embodiment, please refer to Figure 4 , microservices also include department triage microservices, unified payment microservices and operation and maintenance services. After the patient obtains the appointment outpatient authorization token through the self-service terminal device, the token authorization and authentication center of the intelligent gateway stores the authorization log record. Patients can log in to any self-service machine in the hospital for authentication. After passing, they can access a variety of microservice outpatient services, allowing patients who have obtained the authorization token to access the department triage microservice and unified payment microservice through any terminal. The authorization token includes information such as the patient's name, ID number, patient visit ID, department outpatient number, token generation time, limited period and encrypted signature. The token authorization authentication module of the intelligent gateway authorizes the token to contain at least outpatient number, name, department and other data information.

[0028] For more details, please refer to Figure 3 The patient client logs in to the hospital's on-site self-service machine or mobile APP to send an appointment HTTP request to the smart gateway. After the gateway receives the client's HTTP request, it forwards it to the token authorization and authentication center of the smart gateway (such as Figure 2The request authentication module 220) requests authorization, and the token authorization authentication center parses the request and searches the log to determine whether the request has an authorization token. If there is a token, access is allowed; if there is no token, the request is processed by the processing center. The processing center receives the request and identifies the microservice content. The microservice server and the microservice database are accessed through routing to retrieve outpatient data. The processing center analysis module compares the actual number of outpatient visits of the patient department on that day with the preset threshold, determines whether the request is true or false, and returns the approval result to the intelligent gateway token authorization authentication center. The token authentication center decides whether to grant or not grant the token based on the true or false result, and responds to the client's HTTP request through the gateway port.

[0029] In one embodiment, the processing center 200 establishes a department patient waiting time prediction model to predict the patient's department triage time of the day. The department patient waiting time prediction model uses an LSTM neural network algorithm model, inputs the department's appointment volume of the day to predict the patient's visit time of the day.

[0030] In one embodiment, the outpatient volume threshold for the day also includes the second outpatient volume threshold for the day; the second outpatient volume threshold for the day is greater than the first outpatient volume threshold for the day (such as setting m = 1.2 times the predicted outpatient volume of the department); if the outpatient volume scheduled for the day of the department is lower than the first threshold of the outpatient volume for the day, a normal authorization token is sent; if the outpatient volume scheduled for the day of the department exceeds the threshold but does not exceed the second threshold of the outpatient volume for the day, a message is sent to determine whether it is an emergency and whether the emergency surcharge is agreed to be paid. If feedback is received that it is not an emergency or that the emergency surcharge is not agreed to be paid, a non-authorization token is sent, and a notification message that the department outpatient volume for the day has been full is sent to the information feedback module 230 of the intelligent gateway, and the information feedback module 230 sends the notification message to the client (collection terminal 100); if feedback is received that it is an emergency and the emergency surcharge is agreed to be paid, a yellow emergency authorization token is sent.

[0031] In one embodiment, it is set that if the patient reports within M hours before the consultation time, the ordinary token is valid. If the patient reports after M hours before the consultation time, the ordinary token will be invalid or cancelled; the interval selected by the M value is 0.75 to 2 (i.e. 45min to 120min). The M value can be selected according to the actual needs of the hospital, such as 2 hours. In addition, the expired token or the cancelled token cannot be restored. In addition to the ordinary authorization token information, the yellow emergency authorization token is also required to report to the triage department by scanning the code within N hours in advance, otherwise the yellow emergency token will be cancelled. Optionally, the cancelled yellow emergency token cannot be restored. The interval selected by the N value is 0.5 to 0.75 (i.e. 30min to 45min). The M value can be selected according to the actual needs of the hospital, such as 0.5 hours.

[0032] In one embodiment, a priority value is set for each common authorization token according to the proximity between the consultation time and the actual time of the day, and the tokens are sorted according to the priority value to form a M-hour (2-hour) time sequence of patients who are waiting for the triage department to see the doctor. a is a constant greater than 1, t0 is the time of consultation predicted by the analysis module through the waiting time prediction model of patients in the department, and t is the actual time of the day. t0-t represents the closeness between the consultation time and the actual time of the day. The closer the time is, the greater the priority value w.

[0033] In one embodiment, the priority value of the emergency authorization token is set to be β times the priority value of the ordinary authorization token, that is, According to the priority value, patients are added to the outpatient sequence for queuing. In practical applications, the priority value can be changed according to the severity of the emergency disease.

[0034] In one embodiment, the number of emergency authorization tokens for the day is set to N. If the number of emergency appointments in the unified appointment microservice request from the terminal on the day exceeds N, the emergency authorization token will not be granted and the emergency appointment channel will be closed.

[0035] In one embodiment, according to the number of invalidated tokens, the corresponding number of missed appointments is recorded, and the corresponding number of reserved numbers is set; if the reserved number of numbers is greater than or equal to 1, a number request approval service is also provided when an unauthorized token is sent to the terminal. More specifically, if the processing center feedbacks that the token cannot be granted, the intelligent gateway can also send the patient department's missed appointment reserve to the number pool center of the hospital scheduling and calling system. If the current reserve flow is greater than 1, a grant token is returned, and the department's number reserve counter is automatically reduced by 1. If the current reserve flow is less than 1, the message "Today's department outpatient clinic is full, please make an appointment on another day!" is returned.

[0036] In one embodiment, if it is found that the patient's visit time is one hour before the department's off-duty time on that day, the outpatient volume threshold for that day is automatically changed to the actual number of triages on that day. For example, if it is monitored that the patient's visit time is one hour before the hospital's off-duty time on that working day, the threshold parameter setting of the trigger analysis module 320 automatically changes the outpatient volume judgment threshold to the actual number of triages per day to limit ordinary outpatient appointments.

[0037] In one embodiment, please refer to Figure 5 , the analysis and processing flow of the analysis module 320 is as follows:

[0038] When the analysis module 320 receives a patient's outpatient appointment service request for a certain department from the evaluation receiving module of the processing center, it first obtains the outpatient volume threshold of the department on that day through the built-in department triage volume LSTM prediction model of the analysis module 320, and makes a judgment on the current actual appointment volume threshold of the day; if the actual appointment volume at that time has exceeded the threshold, the token grant is not passed, and it is sent to the request authentication module 220 of the intelligent gateway 200 through the evaluation result 330 sending module as token grant information; if the actual appointment volume of the patient user at that time did not exceed the threshold, the patient's waiting time is further obtained through the department patient waiting time prediction model built into the analysis module 320, and then it is determined whether the patient's estimated consultation time is within the doctor's working hours. If so, the token grant is passed, and it is sent to the request authentication module 220 of the intelligent gateway 200 as token grant information through the evaluation result 330 sending module. If not, within the deadline, the token grant statistic for the day is W, and the threshold is changed to W.

[0039] After the patient successfully obtains the department visit microservice appointment at the self-service terminal or mobile APP, the token is stored in the browser cookies, and the token is automatically included each time the service is requested repeatedly. At the same time, the smart gateway automatically stores the token authorization log for verification.

[0040] In some embodiments, the data access method of the artificial intelligence hospital outpatient self-service system also includes daily traffic statistics, which are set in each microservice; among them, for the appointment microservice, the outpatient appointments of each department as of the day are counted; for the department triage microservice, the number of visits completed by each department as of the day is counted; for the unified payment microservice, the summary data of the payment log as of the day is counted; for the operation and maintenance microservice, the terminal operation fault log data as of the day is counted. Each microservice sends the statistically obtained data to the background database in real time or regularly and stores it. The processing center monitors the department triage volume, payment fees, self-service equipment operation and other data of the day in real time through the network. For the unified payment microservice described above, if an abnormal payment fee occurs, the processing center queries the payment log and database data. For the operation and maintenance microservice described above, the system collects the hospital service terminal / mobile service terminal operation status log data through the intelligent gateway. Each terminal device sends the 24-hour hospital self-service terminal operation log data through the intelligent gateway regularly (at 12:00 or 17:00 every day). All log data is stored in the database of the operation and maintenance microservice.

[0041] Optionally, the processing center analysis module can establish a GCN graph neural network terminal log anomaly monitoring model. The model communicates through the intelligent gateway, routes to the operation and maintenance microservice and the operation and maintenance log database, and obtains the self-service terminal device log data for anomaly prediction.

[0042] The present invention provides an artificial intelligence hospital outpatient self-service system data access device of an embodiment, the device comprising: a data receiving module, used to receive a microservice acquisition request sent by a terminal user; a request authentication module, used to parse the microservice acquisition request, query whether the user has an authorization token, and authorize or not authorize the user to obtain the microservice according to the token situation; an evaluation and analysis module, used to query the corresponding database and perform evaluation and analysis according to the content of the microservice request when the user does not have an authorization token, and obtain an evaluation result; an evaluation result sending module, used to give an authorization or non-authorization token according to the evaluation and analysis result and send it to the terminal via an intelligent gateway; wherein the microservice at least includes a unified reservation microservice, when receiving the microservice acquisition request sent by the terminal user, the microservice acquisition request is parsed, the user is queried whether the authorization token is, if there is no authorization token, the corresponding department outpatient clinic is acquired according to the content of the unified reservation microservice request, the outpatient volume threshold and the outpatient volume of the day of the department outpatient clinic are queried, and the outpatient volume of the day of the day of the outpatient volume threshold and the outpatient volume of the day of the day of the outpatient volume are evaluated and analyzed, and the authorization or non-authorization token is obtained and sent to the terminal; the outpatient volume of the day is obtained by establishing a department outpatient volume prediction model.

[0043] In a more preferred embodiment, the data access device of the artificial intelligence hospital outpatient self-service system is not limited to the modules of the above embodiments. Figure 2 , mainly composed of a collection terminal 100, an intelligent gateway 200, and a processing center 300, wherein the collection terminal 100 mainly includes a data collection module 110 and a request sending module 120; the intelligent gateway 200 mainly includes a data receiving module 210, a request authentication module 220, and an information reverse module 230; the processing center 300 includes an evaluation receiving module 310, an analysis module 320, and an evaluation result sending module 330. Preferably, in this embodiment, the client can communicate with multiple intelligent gateways, and the proxy server Nginx between the two is used to implement load balancing. Specifically as follows:

[0044] The data collection terminal 100 such as the client mobile APP and the self-service terminal is equipped with a data collection function. These data collection terminals include a data collection module 110 and a request sending module 120. The data collection module 110 has an HTML web page data input function. As a preference, it can have a voice dialogue input function. The request sending module 120 uses the WebSocket protocol to exchange data between the client and the proxy server Nginx, which is forwarded to the intelligent gateway by the proxy server, or the client and the gateway directly transmit data through the WebSocket protocol.

[0045] The intelligent gateway 200 includes a data receiving module 210, a request authentication module 220, and an information feedback module 230. The data receiving module 210 parses the data according to the microservice corresponding to the microservice ID set in the routing table, and determines the microservice ID ****1 required by the data.

[0046] The data receiving module 210 sends an authorization request of microservice ID****1 to the request authentication module. The request authentication module 220 requests a queue according to microservice ID****1 and sends a load evaluation message of microservice ID****1 to the processing center.

[0047] After the evaluation receiving module 310 of the processing center 300 receives the data request, it forms a microservice ID ****1 request processing queue and sends it to the analysis module 320.

[0048] After receiving the request, the analysis module 320 parses the request information. Set the routing path to the microservice and the corresponding service database, retrieve the data back to the processing center, and start the built-in algorithm to calculate the evaluation result to determine whether the token is authorized and whether the permission is True or False. The permission information is supplemented with the threshold data and the current actual outpatient volume data and other evaluation results, which are sent to the evaluation result sending module 330.

[0049] The evaluation result sending module 330 sends the evaluation result to the request authentication module 220. The request authentication module 220 decides whether to grant token authorization and notifies the data collection client through the information feedback module 230.

[0050] The analysis module of the processing center stores AI algorithms, including outpatient volume threshold parameter settings, outpatient time series prediction LSTM algorithm model, and department waiting time LSTM prediction algorithm model.

[0051] The outpatient volume threshold setting has the function of changing the first threshold and the second threshold, and compares the current triage volume with the threshold, and transmits the threshold and the comparison result to the intelligent gateway for the authentication module to make a token authorization judgment.

[0052] The outpatient volume forecast for the day is predicted by the outpatient time series prediction LSTM neural network algorithm model based on the historical department outpatient data stream, and the department outpatient volume for the next day or week can be obtained. The department outpatient volume determines the medical staff's shift schedule for the next day or week.

[0053] The LSTM neural network algorithm of the department waiting time prediction algorithm selects the historical time records of patients at the self-service check-in machine in the triage department in the past month, and establishes a relationship model with the actual consultation time.

[0054] Preprocess the historical data of the past month. Fill in the missing items and arrange the time according to every 60 minutes of the hour, such as 8:00, 8:01, 8:02, ..., 8:59. If there is no patient triage report event in a certain time, the actual consultation time is filled with 0.

[0055] The department's general outpatient clinic hours are 8 hours from 8:00 to 17:00 every day; the emergency department hours are 24 hours.

[0056] An LSTM neural model was established to predict the triage department visit time of each patient in the next day based on the historical time in the past month.

[0057] The training set is the historical data of department triage within 20 days in a month, and the rest of the data is the test set. The number of network layers is 3. The number of neurons in the input layer is set to M1, the number of neurons in the hidden layer is set to M2, and the number of neurons in the output layer is set to 1. The input vector is 60×8=480 minutes of time points in a day for general outpatient clinics; 60×24=1440 minutes of time points for emergency departments. Offline neural network model training is performed.

[0058] After the offline model training is completed, the patient triage waiting data of the day is input in real time, and the patient department triage visit time is output in real time. According to the priority authority of general outpatient clinics and the priority authority of emergency outpatient clinics, the visit time points are reallocated in the order of authority values, and the visit time queue is corrected.

[0059] If it is monitored that the patient's visit time is 1 hour before the hospital's working day off time, the threshold parameter setting of the trigger analysis module will automatically change the outpatient volume judgment threshold to the actual daily triage volume to limit ordinary outpatient appointments.

[0060] The present invention can achieve the following beneficial technical effects through the above embodiments:

[0061] 1. The outpatient time series prediction model of the processing center is used to predict the outpatient volume of the day. Setting the outpatient volume threshold can effectively control the number of ordinary appointment registrations in hospital departments, and notify patients of the day's triage status in real time to avoid blind medical treatment.

[0062] 2. The waiting time prediction model of the processing center is used to predict the patient's department triage time of the day, and the patient is informed in time to avoid blindly waiting for treatment, truly realize orderly treatment, and improve patient satisfaction.

[0063] 3. By changing the threshold in real time, the number of general outpatient visits on the day can be appropriately limited, the number of emergency outpatient visits can be increased, the medical and nursing human resources can be rationally allocated, and the hospital's treatment capacity can be improved.

[0064] In summary, the implementation of the present invention can effectively adjust the outpatient volume on the client side and avoid unnecessary load on the hospital's backend system. Moreover, the present invention is the first to unify the adjustment of ordinary outpatient appointments and emergency appointments on the same client side, effectively ensuring the normal operation of the system, thereby improving the intelligence level of the hospital's outpatient services, achieving orderly triage and treatment, and reducing patients' blind waiting time and doctors' workload.

[0065] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the above-mentioned embodiments of the methods. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0066] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0067] The above-described embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the present invention. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the attached claims.

Claims

1. A data access method for an artificial intelligence hospital outpatient self-service system, characterized in that: include: Receive a service acquisition request sent by a terminal user; Parse the request for obtaining the service, check whether the user has an authorization token, and authorize or not authorize the user to obtain the service based on the token status; When the user does not have an authorization token, the corresponding database is queried and evaluated and analyzed according to the content of the service request to obtain the evaluation results; Give authorization or non-authorization token according to the evaluation and analysis results and send it to the terminal; Among them, the service at least includes a unified appointment service. When a request for obtaining an appointment service is received from a terminal user, the service request is parsed and a query is made as to whether the user has an authorization token. If there is no authorization token, the corresponding department outpatient clinic is obtained according to the content of the unified appointment service request, and the daily outpatient volume threshold and the daily scheduled outpatient volume of the department outpatient clinic are queried. An evaluation and analysis is performed based on the daily outpatient volume threshold and the daily scheduled outpatient volume, and an authorization or non-authorization token is obtained and sent to the terminal; the daily outpatient volume is obtained by establishing a department outpatient volume prediction model.

2. The data access method of the artificial intelligence hospital outpatient self-service system according to claim 1 is characterized in that: The outpatient volume threshold for the day includes the first outpatient volume threshold for the day, which is the predicted outpatient volume for the day; an evaluation and analysis is performed based on the outpatient volume threshold for the day and the scheduled outpatient volume for the day, and an authorization or non-authorization token is obtained and sent to the terminal, specifically including: if the scheduled outpatient volume for the day of the department is lower than the first outpatient volume threshold for the day, an authorization token is sent; if the scheduled outpatient volume for the day of the department exceeds the threshold, a non-authorization token is sent, and a notification message is sent that the department outpatient volume for the day has been full.

3. The data access method of the artificial intelligence hospital outpatient self-service system according to claim 2 is characterized in that: The outpatient volume threshold for the day also includes the second outpatient volume threshold for the day; the second outpatient volume threshold for the day is greater than the first outpatient volume threshold for the day; an evaluation and analysis is performed based on the outpatient volume threshold for the day and the scheduled outpatient volume for the day, and an authorization or non-authorization token is obtained and sent to the terminal, specifically including: if the scheduled outpatient volume for the day of the department is lower than the first threshold of the outpatient volume for the day, a normal authorization token is sent; if the scheduled outpatient volume for the day of the department exceeds the threshold but does not exceed the second threshold of the outpatient volume for the day, a message is sent indicating whether it is an emergency and whether the emergency surcharge is agreed to be paid. If feedback is received that it is not an emergency or that the emergency surcharge is not agreed to be paid, a non-authorization token is sent, and a notification message that the department outpatient volume for the day has been full is sent; if feedback is received that it is an emergency and the emergency surcharge is agreed to be paid, an emergency authorization token is sent.

4. The data access method of the artificial intelligence hospital outpatient self-service system according to claim 3 is characterized in that: The acquisition of the predicted outpatient volume on the same day specifically includes the following steps: Contact the corresponding database to obtain the department's outpatient history data; Establish a departmental outpatient volume prediction model; Predict the department's outpatient volume for the day; Establish a prediction model for the waiting time of patients in the department; the department outpatient volume prediction model adopts an LSTM neural network algorithm model, inputs the historical outpatient data of the department's patients in the past year, and predicts the department's triage outpatient volume for one day or one week; the department patient waiting time prediction model adopts an LSTM neural network algorithm model, inputs the department's scheduled outpatient volume on the day to predict the patient's visit time on the day.

5. The data access method of the artificial intelligence hospital outpatient self-service system according to claim 3 is characterized in that: For those who are granted ordinary authorization tokens, if the patient reports within M hours before the consultation time, the ordinary authorization token is valid, and if the patient reports M hours before the consultation time, the ordinary authorization token is invalid; the interval selected for the M value is 0.75 to 2; for those who are granted emergency authorization tokens, if the patient reports within N hours before the consultation time, the emergency authorization token is valid, and if the patient reports N hours before the consultation time, the emergency authorization is invalid; The N value is selected in the range of 0.5 to 0.

75.

6. The data access method of the artificial intelligence hospital outpatient self-service system according to claim 5 is characterized in that: Set priority values ​​for each common authorization token based on how close the consultation time is to the actual time of the day a is a constant greater than 1, t0 is the time of consultation predicted by the analysis module through the waiting time prediction model of patients in the department, t is the actual time of the day, t0-t represents the closeness between the consultation time and the actual time of the day, the closer the time is, the greater the priority value w; sort by the size of the priority value to form a time series of M-hour consultation patients, waiting for the triage department to visit; set the priority value of the emergency authorization token to be β times the priority value of the ordinary authorization token, According to the priority value, patients are added to the outpatient sequence and queued for treatment.

7. The data access method of the artificial intelligence hospital outpatient self-service system according to any one of claims 1 to 6, characterized in that: If it is found that the patient's consultation time is close to or one hour before the hospital department's closing time on that day, the outpatient volume threshold for that day will be automatically changed to the actual triage volume on that day.

8. The data access method of the artificial intelligence hospital outpatient self-service system according to any one of claims 3 to 6, characterized in that: The number of emergency authorization tokens for the day is set to N. If the number of emergency appointments in the unified appointment service request from the terminal on the day exceeds N, the emergency authorization token will not be granted and the emergency appointment channel will be closed; According to the number of invalidated tokens, the corresponding number of no-shows is recorded and the corresponding number of reserved numbers is set; if the reserved number of numbers is greater than or equal to 1, the number request approval service is also provided when an unauthorized token is sent to the terminal.

9. The data access method of the artificial intelligence hospital outpatient self-service system according to claim 1 is characterized in that: The services also include department triage services, unified payment services and operation and maintenance services; patients who have obtained authorization tokens are allowed to access department triage services and unified payment services through any terminal; and daily traffic statistics are also included; For appointment services, statistics are collected on the number of outpatient appointments in each department up to that day; For department triage services, statistics are collected on the number of consultations completed by each department as of that day; For unified payment services, the summary data of payment logs up to the current day is collected; For operation and maintenance services, statistics are collected on terminal operation fault log data up to the current day; Send the statistical data to the backend database in real time or regularly and store them; It also includes real-time monitoring of the operating data of each service. For unified payment services, if an abnormal payment occurs, the payment log and database data will be queried; for operation and maintenance services, terminal operation log data will be collected; and by establishing a GCN graph neural network terminal log anomaly monitoring model, the terminal log data will be monitored for anomalies.

10. An artificial intelligence hospital outpatient self-service system data access device, using the method described in any one of claims 1 to 15 above, characterized in that: include: A data receiving module, used to receive a service acquisition request sent by a terminal user; The request authentication module is used to parse the request for obtaining the service, check whether the user has an authorization token, and authorize or not authorize the user to obtain the service based on the token status; The evaluation and analysis module is used to query the corresponding database and perform evaluation and analysis according to the content of the service request to obtain the evaluation result when the user does not have an authorization token; An evaluation result sending module, used for sending a request authentication module according to the evaluation result; An information feedback module is used to feed back the authorization and non-granting status of the authentication module token to the terminal; Among them, the service at least includes a unified appointment service. When a request for obtaining an appointment service is received from a terminal user, the service request is parsed and a query is made as to whether the user has an authorization token. If there is no authorization token, the corresponding department outpatient clinic is obtained according to the content of the unified appointment service request, and the daily outpatient volume threshold and the daily scheduled outpatient volume of the department outpatient clinic are queried. An evaluation and analysis is performed based on the daily outpatient volume threshold and the daily scheduled outpatient volume, and an authorization or non-authorization token is obtained and sent to the terminal; the daily outpatient volume is obtained by establishing a department outpatient volume prediction model.

Images