Registration doctor-seeing analysis scheduling method
By analyzing the order of patient check-in in real time at the check-in terminal, differentiating patient types and optimizing the medical process, the problem of uncertain waiting time for medical treatment is solved, and the efficiency of medical treatment and patient experience are improved.
Patent Information
- Application Number
- CN202511301948.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-12
- Publication Date
- 2026-02-03
AI Technical Summary
In the current medical institution's consultation process, patients are in a passive waiting state at each check-in stage, and the waiting time is highly uncertain, resulting in low consultation efficiency and poor patient experience.
Design a registration and medical visit analysis and scheduling method. By analyzing the patient's check-in order in real time through the check-in terminal, distinguishing between first-time, follow-up and return patients, and combining historical medical records and doctor's consultation time, optimize the check-in order to achieve intelligent scheduling.
By using multidimensional factor analysis, we can reduce patient waiting time, improve consultation efficiency, enhance patient experience, and reduce overcrowding and disorder in the consultation area.
Smart Images
Figure CN121460092A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The application relates to a registration and treatment analysis scheduling method and belongs to the technical field of intelligent treatment. BACKGROUND
[0002] With the continuous improvement of medical informatization level, existing medical institutions generally adopt the combination of online appointment registration and offline check-in and calling to manage the treatment process. Patients usually complete registration in advance through a mobile application or a website, and then arrive at the hospital on the day of treatment, perform initial check-in at the check-in equipment (such as a self-service machine or a triage station) of the corresponding department to confirm the presence and enter the waiting queue, and then go to the corresponding medical technology department to complete the examination after the doctor receives the patient and issues the examination and testing items. After obtaining the report, the patient returns to the examination room area, performs check-in again (i.e., "back-to-examination check-in"), and reenters the waiting queue to wait for the doctor to call again and interpret the report or perform subsequent diagnosis.
[0003] However, in the above treatment process, the patient is in a passive waiting state after completing each check-in link. After initial check-in, the patient needs to wait for the doctor to call, and after completing the examination, the patient needs to wait for the doctor to call again after back-to-examination check-in. The waiting process lacks an effective state synchronization and resource coordination mechanism, and the waiting time has great uncertainty, which often leads to multiple back-and-forth movements between the examination room and the examination department and a long stay time. The above process has the following significant deficiencies:
[0004] (1) Long waiting time and low overall treatment efficiency: after check-in at each link, the patient can only passively wait for the call, and cannot predict or optimize the waiting time. In particular, after back-to-examination check-in, the doctor may still be receiving other patients, resulting in a long stay time for the patient, and the overall treatment process is seriously lengthened.
[0005] (2) Poor patient experience: multiple ineffective waiting and back-and-forth movements not only increase the physical and mental burden of the patient, but also easily lead to congestion and disorder in the examination area, affecting the quality of medical services. SUMMARY
[0006] The technical problem to be solved by the application is to provide a registration and treatment analysis scheduling method, which triggers the execution of patient analysis design based on treatment check-in actions, and systematically and intelligently schedules each treatment link of the patient, thereby improving the overall treatment efficiency.
[0007] The application adopts the following technical scheme to solve the above technical problems: the application designs a registration and treatment analysis and scheduling method, which is based on various check-in terminals arranged in various departments of a hospital and initial check-ins of patients in a target doctor's on-site treatment period on the day of registration and check-in on the check-in terminal according to the registration and the on-site treatment period on the day of registration, and in the case that the number of patients in the on-site treatment check-in sequence corresponding to the target doctor's on-site treatment period on the day is greater than 0, the following steps are executed in real time to update the on-site treatment check-in sequence, and the system performs sequential number calling for treatment according to the on-site treatment check-in sequence;
[0008] Step A. Based on the newly added initial check-in and the newly added return visit check-in after all the examination items issued to the patient corresponding to the initial check-in are completed, the last sequence in the on-site treatment check-in sequence is updated, and for each patient from the next patient to the last patient of the on-site treatment patient at the current time, the initial visit patient, the return visit patient and the patient after examination are distinguished, and then step B is entered;
[0009] Step B. The patients from the last patient to the preset proportion of patients in the on-site treatment check-in sequence are analyzed respectively, if the patient is an initial visit patient or a return visit patient, the historical treatment record is contacted, the first-level analysis and scheduling is performed on the patient, and the on-site treatment check-in sequence is updated; if the patient is a patient after examination, no further processing is performed.
[0010] As a preferred technical scheme of the application: based on the real-time record of the single on-site treatment time of the target doctor for each patient, the average single on-site treatment time t of the target doctor is obtained in real time, after the on-site treatment check-in sequence is updated in step A, according to the number N of patients from the next patient to the last patient of the on-site treatment patient at the current time in the on-site treatment check-in sequence, T=t*N is obtained, the predicted waiting time T of the last patient in the on-site treatment check-in sequence is obtained, and it is judged whether T is greater than the upper limit value of the preset waiting time, if yes, the initial visit patient, the return visit patient and the patient after examination are distinguished for each patient from the next patient to the last patient of the on-site treatment patient at the current time in the on-site treatment check-in sequence, and step B is entered; otherwise, no further processing is performed.
[0011] As a preferred technical scheme of the application: in step A, for each patient from the next patient to the last patient of the on-site treatment patient at the current time in the on-site treatment check-in sequence, the following steps A1 to A3 are performed to distinguish the initial visit patient, the return visit patient and the patient after examination;
[0012] Step A1. It is judged whether the patient has completed all the examination items issued in the department where the target doctor is located, and whether the patient has not yet performed return visit after completion, if yes, the patient is defined as a patient after examination; otherwise, the patient has not yet issued examination items on the day, and step A2 is entered;
[0013] Step A2. Determine whether the patient has a completed historical medical record before the current day in the department of the target doctor. If yes, go to step A3; otherwise, define the patient as a first-visit patient;
[0014] Step A3. Determine whether the patient has a follow-up medical order in the most recent completed historical medical record in the department of the target doctor and whether the current day meets the preset date fluctuation range corresponding to the follow-up medical order date. If yes, define the patient as a follow-up patient; otherwise, define the patient as a first-visit patient.
[0015] As a preferred technical solution of the present application: in step B, a predetermined number of patients are analyzed from the last patient in the medical appointment sign-in sequence to the front , if the patient is a first-visit patient or a follow-up patient, steps B-1 to B-5 are performed for first-level analysis and scheduling of the patient, and the medical appointment sign-in sequence is updated, where N represents the number of patients from the next patient to the last patient in the medical appointment sign-in sequence at the current time, 0 , a represents the preset proportion,
[0016] Step B-1. If the patient is a first-visit patient, go to step B-2; if the patient is a follow-up patient, go to step B-5.
[0017] Step B-2. Obtain the disease data of the first-visit patient in the preset first-visit questions of the department of the target doctor, form the first-visit disease data corresponding to the first-visit patient, and go to step B-3.
[0018] Step B-3. Based on each completed historical medical record P of the target doctor treating patients whose age is within the preset fluctuation range of the age of the first-visit patient, obtain the historical disease data p in each completed historical medical record P, and obtain the similarity p' between the first-visit disease data and each historical disease data p, and then go to step B-4.
[0019] Step B-4. Determine whether there is historical disease data p with a similarity p' greater than a preset first similarity threshold. If yes, obtain the intersection of the examination item sets in the completed historical medical record P to which each historical disease data p with a similarity p' greater than the preset first similarity threshold belongs, and issue each examination item in the intersection to the first-visit patient, and update the first-visit patient in the medical appointment sign-in sequence; otherwise, do not make any further processing, and end the first-level analysis and scheduling of the first-visit patient.
[0020] Step B-5. Obtain each examination item in the most recent completed historical medical record of the follow-up patient in the department of the target doctor, and issue each examination item to the follow-up patient based thereon, and update the follow-up patient in the medical appointment sign-in sequence.
[0021] As a preferred technical solution of the present application: further comprising step C as follows, after the patients in the preset proportion from the last patient in the clinic sign-in sequence are analyzed, entering step C;
[0022] Step C. According to the number N of patients from the next patient to the last patient in the clinic sign-in sequence at the current time, obtain the predicted waiting time T of the last patient in the clinic sign-in sequence according to T=t*N, and determine whether T is greater than the upper limit value of the preset waiting time, yes, enter step D; otherwise, do not make any further processing;
[0023] Step D. Analyze the patients in the preset proportion from the last patient in the clinic sign-in sequence, if the patient is a first visit patient, contact the completed historical medical record, perform secondary analysis scheduling for the patient, update the clinic sign-in sequence, if the patient is a post-examination return visit patient, do not make any processing.
[0024] As a preferred technical solution of the present application: in step D, the patients in the preset proportion from the last patient in the clinic sign-in sequence are analyzed, if the patient is a first visit patient, the following steps D-1 to D-6 are performed, the secondary analysis scheduling is performed for the patient, and the clinic sign-in sequence is updated, wherein 0<b<1, b represents the preset proportion;
[0025] Step D-1. Obtain the disease data of the first visit patient in the preset first visit question of the department where the target doctor is located, to form the first visit disease data corresponding to the first visit patient, and enter step D-2;
[0026] Step D-2. Based on the completed historical medical records P of the target doctor treating patients whose age is within the preset fluctuation range of the age of the first visit patient, obtain the historical disease data p in each completed historical medical record P, and obtain the similarity p' between the first visit disease data and each historical disease data p, and then enter step D-3;
[0027] Step D-3. Determine whether there is historical disease data p with similarity p' greater than the preset second similarity threshold, yes, obtain the intersection of the inspection item sets in the completed historical medical record P to which each historical disease data p with similarity p' greater than the preset second similarity threshold belongs, to form the main reference inspection item set, and then enter step D-4; otherwise, directly enter step D-4; the preset second similarity threshold is less than the preset first similarity threshold;
[0028] Step D-4. Based on other doctors in the department of the target doctor whose job level is not lower than that of the target doctor, obtain each completed historical medical record Q of patients treated by the other doctors whose age is within the preset fluctuation range of the age of the initial diagnosis patient, and obtain the historical disease data q in each completed historical medical record Q, then obtain the similarity q' between the initial diagnosis disease data and each historical disease data q, and then go to step D-5;
[0029] Step D-5. Determine whether there is historical disease data q with similarity q' greater than the preset first similarity threshold, yes, obtain the intersection of the inspection item set in each completed historical medical record Q to which each historical disease data q with similarity q' greater than the preset first similarity threshold belongs, to form an auxiliary reference inspection item set, and then go to step D-6; otherwise, go to step D-6 directly.
[0030] Step D-6. If there is a main reference inspection item set and an auxiliary reference inspection item set, obtain the union of the main reference inspection item set and the auxiliary reference inspection item set, and issue each inspection item in the union to the initial diagnosis patient, and update the initial diagnosis patient in the medical record sign-in sequence.
[0031] If there is only an auxiliary reference inspection item set, issue each inspection item in the auxiliary reference inspection item set to the initial diagnosis patient, and update the initial diagnosis patient in the medical record sign-in sequence.
[0032] If there is only a main reference inspection item set, dispatch other idle doctors in the department of the target doctor, issue each inspection item in the main reference inspection item set to the initial diagnosis patient, and update the initial diagnosis patient in the medical record sign-in sequence.
[0033] If neither the main reference inspection item set nor the auxiliary reference inspection item set exists, no further processing is performed, and the secondary analysis scheduling for the initial diagnosis patient is ended.
[0034] As a preferred technical solution of the present application, the similarity between the initial diagnosis disease data and the historical disease data is obtained as follows.
[0035] First, obtain each disease word in the initial diagnosis disease data corresponding to the preset disease word set, and obtain each disease word in the historical disease data corresponding to the preset disease word set according to the preset disease word set combined with the conversion of synonyms and antonyms; then obtain the disease vector corresponding to each disease word in the initial diagnosis disease data and the disease vector corresponding to each disease word in the historical disease data through Word2Vec; finally, calculate the distance m between the disease vector of the initial diagnosis disease data and the disease vector of the historical disease data through the preset vector distance method, and take The similarity between the initial diagnosis data and each historical diagnosis data.
[0036] As a preferred technical solution of the application: the preset vector distance mode is a sine distance or a cosine distance.
[0037] As a preferred technical solution of the application: the department pre-sets each initial diagnosis question, including disease and duration.
[0038] As a preferred technical solution of the application: for the registration of the target doctor's office hours, when the patient successfully registers for the target doctor's medical number, a large language model dialogue window with the preset medical popular science knowledge graph database of the department where the target doctor is located is pushed to the patient, and a preset duration of the cancellation window period is provided for the patient, the patient realizes related medical popular science knowledge search through the provided large language model dialogue window, and is allowed to cancel the registration within the cancellation window period, and if the registration is cancelled, the empty number caused by the cancellation is re-opened for appointment.
[0039] The registration and diagnosis analysis scheduling method of the application has the following technical effects compared with the prior art by adopting the above technical solution:
[0040] The application designs a registration and diagnosis analysis scheduling method, which analyzes the diagnosis sign-in order of each patient group composed of initial sign-in and re-diagnosis sign-in in real time, analyzes the longest waiting time as a triggering condition, triggers the analysis of the initial diagnosis patients and re-diagnosis patients distinguished by the patient analysis design, contacts the completed historical diagnosis records of each historical patient treated by the target doctor registered by the patient, designs and executes a primary analysis scheduling to update the diagnosis sign-in order, and further triggers the contact of the completed historical diagnosis records of each historical patient treated by each doctor in the department where the target doctor is located, designs and executes a secondary analysis scheduling to update the diagnosis sign-in order, so that the application designs a comprehensive multi-dimensional factor analysis and a progressive two-level analysis scheduling scheme, which can reduce the patient waiting time for diagnosis to the maximum extent while ensuring the quality of diagnosis and improving the efficiency of diagnosis. BRIEF DESCRIPTION OF DRAWINGS
[0041] Figure 1 is a flowchart of the registration and diagnosis analysis scheduling method designed by the application. DETAILED DESCRIPTION
[0042] The specific embodiments of the application will be further described in detail below with reference to the accompanying drawings.
[0043] The application designs a registration and treatment analysis scheduling method. In the design scheme, for the registration of the target doctor's office hours, the patient successfully registers for the target doctor's treatment number, and at the same time, a large language model dialogue window with the target doctor's department preset medical popular science knowledge graph as the background database is pushed to the patient, and a preset time length of the cancellation window period is provided for the patient. The patient can find relevant medical popular science knowledge through the provided large language model dialogue window, and is allowed to cancel the number within the cancellation window period. If the number is cancelled, the empty number caused by the cancellation will be reopened for appointment.
[0044] Here, the design of the cancellation window period adopts a large language model background database that selects the target doctor's department preset medical popular science knowledge graph. That is, in the design of the intelligent dialogue interaction with the patient, a preset limited range of data is used as the basis to ensure the accuracy of the interactive conversation. Therefore, in the registration link, if the patient's medical problem can be solved through the large language model dialogue window, the patient does not need to go to the scene for treatment, and the number source can be given up to leave the limited number source to patients who need it more, thereby improving the overall treatment efficiency of patients from the initial stage.
[0045] In actual application, when the time comes to the target doctor's treatment day, based on the various check-in terminals deployed by each department of the hospital, the patient's registration for the target doctor's office hours on the same day, the initial check-in according to the registration for the same day's office hours on the check-in terminal, and the real-time recording of the target doctor's single office hours for each patient, the average single office hours t of the target doctor is obtained in real time. In actual application, when the number of patients in the target doctor's office hours on the same day is greater than 0, as shown in Figure 1 The following steps A to D are executed in real time to update the treatment check-in sequence, and the system calls the number for treatment according to the treatment check-in sequence.
[0046] Step A. Based on the newly added initial check-in and the newly added return check-in after the patient corresponding to the initial check-in completes all the examination items, the last sequence of the treatment check-in sequence is updated. According to the number N of the next patient to the last patient in the treatment check-in sequence from the current treatment patient, T=t*n is obtained, and the predicted waiting time T of the last patient in the treatment check-in sequence is obtained. That is, the maximum waiting time for the analysis object in the current treatment check-in sequence is judged. If T is greater than the preset upper limit of the waiting time, the patients from the next patient to the last patient in the treatment check-in sequence are executed in the following steps A1 to A3, and the initial diagnosis patient, the return patient, and the examination patient are distinguished, and step B is entered. Otherwise, no further processing is performed.
[0047] Step A1. Determine whether the patient has completed all the prescribed examination items in the department where the target doctor is located and has not had a follow-up visit after completion. If so, define the patient as a patient for follow-up visit after examination; otherwise, that is, if the patient has not had the examination items prescribed on the same day, go to Step A2.
[0048] Step A2. Determine whether there is a historical medical record of the patient's completed visit before the same day in the department where the target doctor is located. If so, go to Step A3; otherwise, define the patient as a first-time patient.
[0049] Step A3. Determine whether there is a follow-up visit order in the historical medical record of the patient's completed visit closest to the same day in the department where the target doctor is located, and whether the same day meets the preset date fluctuation range corresponding to the follow-up visit order date requirement. If so, define the patient as a follow-up patient; otherwise, define the patient as a first-time patient.
[0050] Step B. Analyze a preset number of patients starting from the last patient in the order of visit check-in forward. If the patient is a first-time patient or a follow-up patient, perform the following Steps B-1 to B-5 to perform primary analysis and scheduling for the patient and update the order of visit check-in; if the patient is a patient for follow-up visit after examination, do nothing. Here, N represents the number of patients from the next patient of the current patient for visit check-in to the last patient in the order of visit check-in, 0 < a < 1, a represents the preset ratio, denotes rounding up.
[0051] Step B-1. If the patient is a first-time patient, go to Step B-2; if the patient is a follow-up patient, go to Step B-5.
[0052] Step B-2. Obtain the disease data of the initial diagnosis problems such as diseases and duration preset for the first-time patient corresponding to the target doctor's department, form the initial diagnosis disease data corresponding to the first-time patient, and go to Step B-3.
[0053] Step B-3. Based on the historical medical records P of all patients treated by the target doctor whose ages are within the preset fluctuation range above and below the age of the first-time patient, obtain the historical disease data p in each historical medical record P, and obtain the similarity p' between the initial diagnosis disease data and each historical disease data p, and then go to Step B-4.
[0054] Step B-4. Determine whether there is a historical disease data p with a similarity p' greater than the preset first similarity threshold. If so, obtain the intersection of the sets of examination items in the historical medical records P to which the historical disease data p with a similarity p' greater than the preset first similarity threshold belong, prescribe the examination items in the intersection for the first-time patient, and delete the first-time patient from the order of visit check-in for update; otherwise, do not perform any further processing and end the primary analysis and scheduling for the first-time patient.
[0055] Step B-5. Obtain each examination item of the re-visit patient in the target doctor's department from the nearest one-day completed historical visit record of the re-visit patient, and issue each examination item for the re-visit patient, and update the visit sign-in sequence by deleting the re-visit patient.
[0056] Based on the design of the above steps A to B, after the step B is completed for the preset proportion of patients from the last patient in the visit sign-in sequence, the following step C is entered.
[0057] Step C. According to the number N of patients from the next patient to the last patient of the current time in the visit sign-in sequence, T = t x N is obtained, and the predicted waiting time T of the last patient in the visit sign-in sequence is obtained, and it is judged whether T is greater than the preset upper limit of the waiting time. If yes, step D is entered; otherwise, no further processing is done.
[0058] Step D. Analyze the preset number of patients from the last patient in the visit sign-in sequence, respectively, if the patient is a first visit patient, execute the following steps D-1 to D-6, and perform secondary analysis scheduling for the patient, update the visit sign-in sequence, if the patient is a post-examination re-visit patient, no further processing is done; wherein 0 < b < 1, b represents the preset proportion.
[0059] Step D-1. Obtain the disease data of the first visit patient corresponding to the preset first visit questions such as disease and duration of the target doctor's department, and form the first visit disease data corresponding to the first visit patient, and enter step D-2.
[0060] Step D-2. Based on the completed historical visit records P of the target doctor's patients whose ages are within the preset fluctuation range of the first visit patient's age, obtain the historical disease data p in each completed historical visit record P, and obtain the similarity p' between the first visit disease data and each historical disease data p, and then enter step D-3.
[0061] Step D-3. It is judged whether there is historical disease data p with similarity p' greater than the preset second similarity threshold value. If yes, obtain the intersection of the examination item sets in the completed historical visit records P to which each historical disease data p with similarity p' greater than the preset second similarity threshold value belongs, and form the main reference examination item set, and then enter step D-4; otherwise, directly enter step D-4; the preset second similarity threshold value is less than the preset first similarity threshold value.
[0062] Step D-4. Based on other doctors in the department of the target doctor whose job level is not lower than that of the target doctor, obtain each completed historical visit record Q of patients treated by the other doctors whose age is within the preset fluctuation range of the age of the initial diagnosis patient, and obtain the historical disease data q in each completed historical visit record Q, and then obtain the similarity q' between the initial diagnosis disease data and each historical disease data q, and then go to Step D-5.
[0063] Step D-5. Determine whether there is historical disease data q with similarity q' greater than the preset first similarity threshold. If yes, obtain the intersection between the inspection item sets in each completed historical visit record Q to which each historical disease data q with similarity q' greater than the preset first similarity threshold belongs, to form an auxiliary reference inspection item set, and then go to Step D-6. Otherwise, go to Step D-6 directly.
[0064] Step D-6. The following cases are specifically designed for execution respectively:
[0065] (1) If there are both the main reference inspection item set and the auxiliary reference inspection item set, obtain the union set between the main reference inspection item set and the auxiliary reference inspection item set, and issue each inspection item in the union set to the initial diagnosis patient, and update the initial diagnosis patient in the visit sign-in sequence by deleting the initial diagnosis patient;
[0066] (2) If only the auxiliary reference inspection item set exists, issue each inspection item in the auxiliary reference inspection item set to the initial diagnosis patient, and update the initial diagnosis patient in the visit sign-in sequence by deleting the initial diagnosis patient;
[0067] (3) If only the main reference inspection item set exists, dispatch other idle doctors in the department of the target doctor, and issue each inspection item in the main reference inspection item set to the initial diagnosis patient, and update the initial diagnosis patient in the visit sign-in sequence by deleting the initial diagnosis patient;
[0068] (4) If neither the main reference inspection item set nor the auxiliary reference inspection item set exists, do not make any further processing, and end the secondary analysis scheduling for the initial diagnosis patient.
[0069] The similarity between the initial diagnosis disease data and the historical disease data involved in Steps B-3 and D-2 in the above design scheme is obtained by the following method in actual application:
[0070] First, according to the preset disease word set, combined with the conversion of synonyms and synonyms, obtain each disease word corresponding to the preset disease word set in the initial diagnosis disease data, and obtain each disease word corresponding to the preset disease word set in the historical disease data; Then, through Word2Vec, obtain the disease vector corresponding to each disease word in the initial diagnosis disease data, and obtain the disease vector corresponding to each disease word in the historical disease data; Finally, through the preset method such as sine distance or cosine distance, the distance m between the disease vector of the initial diagnosis disease data and the disease vector of the historical disease data is calculated, and the distance m is taken as The similarity between the initial diagnosis disease data and each historical disease data is obtained.
[0071] The above technical scheme designs a registration and diagnosis analysis scheduling method, which analyzes the initial registration and the registration of each patient group composed of reexamination registration in real time, and analyzes the initial diagnosis and reexamination patients distinguished by the patient analysis design as the triggering condition in sequence according to the longest waiting time, triggers the initial diagnosis and reexamination patients distinguished by the patient analysis design, contacts the completed historical diagnosis records of each historical patient treated by the target doctor registered by the patient, and designs to execute the first-level analysis scheduling to update the registration and diagnosis sequence, and further triggers the contact of each completed historical diagnosis record of each historical patient treated by each doctor in the department of the target doctor, designs to execute the second-level analysis scheduling to update the registration and diagnosis sequence, so that the comprehensive multi-dimensional factor analysis of the design scheme of the present application is realized, and the two-level analysis scheduling scheme is designed in a progressive manner, which can guarantee the quality of diagnosis, minimize the waiting time of patients, and improve the efficiency of diagnosis
[0072] The embodiments of the present application are described in detail above in combination with the drawings, but the present application is not limited to the above-mentioned embodiments, and various changes can be made within the knowledge of those skilled in the art without departing from the purpose of the present application.
Claims
1. A method for analyzing and scheduling registration and medical visits, characterized in that: Based on the various check-in terminals deployed by each department of the hospital and the initial check-in of the patient at the registration and check-in terminal during the target doctor's office hours, and in the case where the number of patients in the target doctor's office hours is greater than 0, the following steps are executed in real time to update the check-in order, and the system performs sequential number calling according to the check-in order: Step A. Based on the newly added initial check-in and the newly added return visit check-in after the patient has completed all the examination items, the last order of the check-in order is updated. For each patient from the next patient to the last patient in the check-in order, the patient is classified as a first visit patient, a return visit patient, and a post-examination return visit patient, and then step B is entered. Step B. For a predetermined number of patients from the last patient in the check-in order, if the patient is a first visit patient or a return visit patient, the patient's historical visit record is contacted for first-level analysis and scheduling, and the check-in order is updated. If the patient is a post-examination return visit patient, no further processing is performed.
2. The method of claim 1, wherein: Based on the real-time recording of the target doctor's single visit time for each patient, the average single visit time t of the target doctor is calculated in real time. After the check-in order is updated in step A, the number N of patients from the next patient to the last patient in the check-in order is obtained, and the predicted waiting time T of the last patient in the check-in order is obtained according to T=t×N. If T is greater than the upper limit of the preset waiting time, the patients from the next patient to the last patient in the check-in order are classified as first visit patients, return visit patients, and post-examination return visit patients, and step B is entered. Otherwise, no further processing is performed.
3. The analysis and dispatch method for registered visits according to claim 1 or 2, characterized in that: In step A, for each patient from the next patient to the last patient in the check-in order, the following steps A1 to A3 are performed to classify the patient as a first visit patient, a return visit patient, or a post-examination return visit patient. Step A1. Determine whether the patient has completed all the examination items in the target doctor's department and has not yet returned for a visit. If yes, the patient is defined as a post-examination return visit patient. Otherwise, the patient has not yet been prescribed examination items, and step A2 is entered. Step A2. Determine whether the patient has a completed historical visit record before the current day in the target doctor's department. If yes, step A3 is entered. Otherwise, the patient is defined as a first visit patient. Step A3. Determine whether the patient has a return visit medical order in the target doctor's department from the last completed historical visit record and whether the current day meets the date requirement of the return visit medical order within the predetermined date fluctuation range. If yes, the patient is defined as a return visit patient. Otherwise, the patient is defined as a first visit patient.
4. The analysis and dispatch method for registered visit according to claim 1 or 2, characterized in that: In step B, the last patient in the registration order is taken as a reference, and the patients before the last patient are taken as a reference The number of patients is analyzed, and if the patient is a new patient or a returning patient, steps B-1 to B-5 are performed to schedule a primary analysis for the patient and update the registration order, where N represents the number of patients from the next patient to the last patient in the registration order at the current time, 0 represents the ceiling function; Step B-1. If the patient is a first visit patient, step B-2 is entered. If the patient is a return visit patient, step B-5 is entered. Step B-2. Obtain the disease data of the initial diagnosis patient in the preset initial diagnosis questions of the department of the target doctor corresponding to the initial diagnosis patient, form the initial diagnosis disease data corresponding to the initial diagnosis patient, and enter step B-3. Step B-3. Obtain the historical disease data p in the completed historical medical records P of the target doctor based on the age of the patient treated by the target doctor within the preset fluctuation range of the age of the initial diagnosis patient, and obtain the similarity p' between the initial diagnosis disease data and each historical disease data p, and then enter step B-4. Step B-4. Determine whether there is historical disease data p with a similarity p' greater than a preset first similarity threshold, yes, obtain the intersection between the examination item sets in the completed historical medical records P to which each historical disease data p with a similarity p' greater than a preset first similarity threshold belongs, and issue each examination item in the intersection to the initial diagnosis patient, and update the initial diagnosis patient in the medical registration sequence; otherwise, no further processing is performed, and the first-level analysis scheduling for the initial diagnosis patient is ended. Step B-5. Obtain the examination items of the re-visit patient in the completed historical medical records of the target doctor on the day closest to the day, and issue each examination item to the re-visit patient based on the examination items, and update the re-visit patient in the medical registration sequence.
5. The method of claim 4, wherein: Further comprising step C as follows, after the step B completes the analysis for the patients in the medical registration sequence from the last patient to the front in a preset proportion, enter step C; Step C. According to the number N of patients from the next patient to the last patient of the current time in the medical registration sequence, obtain the predicted waiting time T of the last patient in the medical registration sequence according to T=t×N, and determine whether T is greater than a preset upper limit of the waiting time, yes, enter step D; otherwise, no further processing is performed; Step D. Analyze the patients in the medical registration sequence from the last patient to the front in a preset proportion, if the patient is an initial diagnosis patient, contact the completed historical medical records, and perform a second-level analysis scheduling for the patient, update the medical registration sequence, if the patient is a post-examination re-visit patient, no further processing is performed.
6. The method of claim 5, wherein: In step D, for the last patient in the clinic sign-in sequence, the number of patients is analyzed, and if the patient is a first-time patient, steps D-1 to D-6 are executed, and the secondary analysis is scheduled for the patient, and the clinic sign-in sequence is updated, where 0<b<1, b represents the preset proportion. A number of patients are analyzed, and if the patient is a first-time patient, steps D-1 to D-6 are executed, and the secondary analysis is scheduled for the patient, and the clinic sign-in sequence is updated, where 0<b<1, b represents the preset proportion. Step D-1. Obtain the disease data of the initial diagnosis patient in the preset initial diagnosis questions of the department of the target doctor corresponding to the initial diagnosis patient, form the initial diagnosis disease data corresponding to the initial diagnosis patient, and enter step D-2; Step D-2. Obtain the historical disease data p in the completed historical medical records P of the target doctor based on the age of the patient treated by the target doctor within the preset fluctuation range of the age of the initial diagnosis patient, and obtain the similarity p' between the initial diagnosis disease data and each historical disease data p, and then enter step D-3; Step D-3. Obtain the historical disease data p in the completed historical medical records P of the target doctor based on the age of the patient treated by the target doctor within the preset fluctuation range of the age of the initial diagnosis patient, and obtain the similarity p' between the initial diagnosis disease data and each historical disease data p, and then enter step D-3; Step D-3. Determine whether there is a similarity p' greater than a preset second similarity threshold. If yes, obtain the intersection between the inspection item sets in the completed historical medical records P to which each historical medical condition data p with a similarity p' greater than a preset second similarity threshold belongs, to form a main reference inspection item set, and then proceed to step D-4; otherwise, proceed directly to step D-4; the preset second similarity threshold is less than the preset first similarity threshold; Step D-4. Based on the other doctors in the department of the target doctor with a job level not lower than that of the target doctor, obtain each completed historical medical record Q of a patient treated by the other doctors, whose age is within a preset fluctuation range of the age of the initial diagnosis patient, and obtain the historical medical condition data q in each completed historical medical record Q, then obtain the similarity q' between the initial diagnosis medical condition data and each historical medical condition data q, and then proceed to step D-5; Step D-5. Determine whether there is a historical medical condition data q with a similarity q' greater than a preset first similarity threshold. If yes, obtain the intersection between the inspection item sets in the completed historical medical records Q to which each historical medical condition data q with a similarity q' greater than a preset first similarity threshold belongs, to form an auxiliary reference inspection item set, and then proceed to step D-6; otherwise, proceed directly to step D-6; Step D-6. If there is a main reference inspection item set and an auxiliary reference inspection item set, obtain the union between the main reference inspection item set and the auxiliary reference inspection item set, and issue each inspection item in the union to the initial diagnosis patient, and update the initial diagnosis patient in the medical record check-in sequence; If there is only an auxiliary reference inspection item set, issue each inspection item in the auxiliary reference inspection item set to the initial diagnosis patient, and update the initial diagnosis patient in the medical record check-in sequence; If there is only a main reference inspection item set, dispatch other idle doctors in the department of the target doctor, issue each inspection item in the main reference inspection item set to the initial diagnosis patient, and update the initial diagnosis patient in the medical record check-in sequence; If neither a main reference inspection item set nor an auxiliary reference inspection item set exists, no further processing is performed, and the secondary analysis scheduling for the initial diagnosis patient is ended.
7. The method of claim 4 or 6, wherein: The similarity between the initial diagnosis medical condition data and the historical medical condition data is obtained as follows; First, according to the preset disease word set, combined with the conversion of synonyms and synonyms, obtain each disease word corresponding to the preset disease word set in the initial diagnosis disease data, and obtain each disease word corresponding to the preset disease word set in the historical disease data; then through Word2Vec, obtain the disease vector corresponding to each disease word in the initial diagnosis disease data, and obtain the disease vector corresponding to each disease word in the historical disease data; finally, through the preset vector distance mode, the distance m between the disease vector of the initial diagnosis disease data and the disease vector of the historical disease data is calculated, and the similarity between the initial diagnosis disease data and each historical disease data is obtained. constitute 8. The method of claim 7, wherein: The preset vector distance method is a sine distance or a cosine distance.
9. The analysis and dispatch method for registered visit according to claim 4 or 6, characterized in that: The department preset initial diagnosis questions include medical conditions and duration.
10. The method of claim 1, wherein: For the registration of the target doctor's office hours, when a patient successfully registers for a target doctor's medical record, a large language model dialogue window with the target doctor's department preset medical knowledge graph as the background database is pushed to the patient, and a preset duration of the number cancellation window is provided for the patient. The patient can find relevant medical knowledge through the provided large language model dialogue window, and is allowed to cancel the number within the number cancellation window. If the number is cancelled, the empty number caused by the cancellation is re-opened for appointment.