AI generation-based outpatient service management method and system

Abstract

The invention belongs to the technical field of outpatient service management, and discloses a generative AI-based outpatient service management method and system, and the method comprises the steps: obtaining the basic information, symptom description and historical medical record data of a patient when the patient sees a doctor through an AI outpatient service management system, and carrying out the semantic understanding and preliminary triage prediction of the symptom of the patient through a generative AI model, and generating possible disease types and corresponding department suggestions. According to the invention, semantic understanding and triage prediction are carried out on the symptoms of the patient through the generative AI, and intelligentization of the appointment process is realized. Patients do not need to manually select cumbersomely, and the system automatically generates department suggestions, so that the appointment time is greatly shortened, and the convenience and efficiency of appointment of the patients are improved. The constructed patient priority evaluation model and the department resource state model enable the calculation of the reservation time to be more scientific and reasonable, coordinate the time arrangement of the reservation and the preliminary diagnosis stage, effectively avoid reservation conflicts and long-time waiting, and improve the overall efficiency and accuracy of outpatient reservation.

Description

Technical Field

[0003] The present invention belongs to the technical field of outpatient management, and specifically relates to an outpatient management method and system based on generative AI. Background Art

[0004] The current outpatient management system has many problems that need to be solved urgently. The patient appointment process is cumbersome, often requiring manual selection of departments and doctors, and it is difficult to obtain the optimal appointment time in real time, resulting in long waiting times for appointments. The triage process lacks a dynamic adjustment mechanism, and traditional triage standards cannot take into account the urgency of the patient's condition and the resource load of the department. It is easy for patients with mild symptoms to wait too long or patients with severe conditions to not be diagnosed and treated in time. In terms of resource scheduling, doctor scheduling and examination equipment allocation mostly rely on manual experience, which cannot accurately match real-time patient traffic, often resulting in the contradiction of idle equipment and patients queuing. In addition, the generation of medical records and treatment recommendations is inefficient, and doctors need to spend a lot of time processing paperwork, which affects the efficiency of diagnosis and treatment. The existing system lacks the ability to deeply mine and predict historical data, making it difficult to achieve forward-looking optimization of outpatient processes and unable to meet the needs of smart healthcare for efficient and precise management.

[0005] To this end, the present invention proposes an outpatient management method and system based on generative AI. Summary of the Invention

[0006] The purpose of the present invention is to provide an outpatient management method and system based on generative AI, which solves the problems existing in the prior art.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] A generative AI-based outpatient management method, comprising:

[0009] When a patient visits a doctor through the AI ​​outpatient management system, the system obtains the patient's basic information, symptom description, and historical medical records. A generative AI model is then used to semantically understand the patient's symptoms and conduct preliminary triage predictions, generating possible disease types and corresponding department recommendations.

[0010] Set the target appointment cycle duration, obtain the current appointment volume of each department, doctor's consultation status and expected treatment time, implement time coordination strategy, combine patient priority and department resources, calculate the optimal appointment time for each patient, and optimize the overall appointment cycle;

[0011] Establish a patient condition severity assessment model, input patient symptoms, vital signs and other data to generate triage priorities, implement resource balancing strategies based on the real-time number of patients in each department and the doctor's consultation progress, and dynamically adjust the patient triage department and sequence;

[0012] Obtain the doctor's professional field, diagnosis and treatment efficiency and current workload, obtain the use status, maintenance cycle and appointment status of the examination equipment, implement resource modeling optimization strategy, calculate the resource requirements of each link in the diagnosis and treatment process, and optimize the resource allocation objective function;

[0013] Generative AI is used to assist doctors in generating differential diagnosis ideas and treatment plan recommendations, record the diagnosis and treatment process in real time and automatically generate structured medical records, and generate follow-up plans and health management recommendations based on the patient's condition;

[0014] Collect outpatient operation data, use generative AI analysis to identify bottleneck problems, optimize appointment scheduling, triage and resource allocation models and strategies, and present the optimization results to medical staff and patients.

[0015] Preferably, the intelligent appointment scheduling includes:

[0016] Build a patient priority assessment model to generate appointment priority weights based on the urgency of the patient's condition, the early or late appointment time, and special needs;

[0017] Establish a department resource status model, including the number of doctors in each department, the reception capacity, the number of patients with appointments, and the expected consultation time;

[0018] Calculate the appointment time for patient i, i = 1, 2, ..., n:

[0019] Reservation start time t1(i) start =max(t1(i-1) end ,t2(i-1) start -T buffer ), where t1(i-1) end The appointment end time for patient i-1, t2(i-1) start is the time when patient i-1 first visited the hospital, T buffer Provide buffer time between appointment and initial consultation;

[0020] Time t2(i) when the first visit begins start =mx(t1(i) end ,t2(i-1) end ), where t1(i) end The appointment end time for patient i, t2(i-1) end The time when the initial visit of patient i-1 ends.

[0021] Preferably, the execution time coordination strategy further includes:

[0022] Limiting the appointment or initial consultation stage of each patient at any time;

[0023] Calculate the appointment duration of the i-th patient T(i) = T1(i) + T2(i), where T1(i) is the appointment phase duration and T2(i) is the initial consultation waiting phase duration;

[0024] Get the target cycle time T target , calculate the duration objective function

[0025] Preferably, the execution resource modeling optimization strategy includes:

[0026] Calculate the resource requirements of department j:

[0027] Doctor resource requirements Among them, P 1,j is the average diagnosis and treatment efficiency of doctor j in department, ) is the time when department j starts to treat patients;

[0028] Check device resource requirements Among them, P 2,j The maximum power used by the device. The time when the device starts to be used, T 2,j The life cycle of the equipment;

[0029] Waiting space resource demand E 3,j (t)=-P 3,j ×T 3,j , where P 3,j is the waiting space unit release rate, T 3,j The average waiting time for patients.

[0030] Preferably, the execution resource modeling optimization strategy further includes:

[0031] Calculate the resource status of department j at any time t: Resource status = Resource release - First resource consumption, where First resource consumption is the sum of doctor resources and equipment resources consumed during the diagnosis and treatment phase, and Resource release is the sum of waiting space resources released during the patient departure phase.

[0032] Calculate the rate of change of resource status within the time step Δt, and define the resource allocation objective function as minimizing the rate of change.

[0033] An outpatient management system based on generative AI, including:

[0034] Patient information modeling module, used to input and extract patients' basic information, symptom descriptions, historical medical records, vital signs and other data;

[0035] Intelligent appointment scheduling module, used to implement appointment process control, including patient priority assessment, department resource status monitoring and appointment time calculation;

[0036] Dynamic triage module, used to calculate triage priority and dynamically adjust the triage order based on the patient's condition and department resources;

[0037] Resource coordination module, used to coordinate resource allocation among departments and patient treatment order according to target cycle duration;

[0038] A diagnosis and treatment assistance module is used to calculate the resource requirements of each department, generate diagnosis and treatment recommendations, and automatically record medical records;

[0039] Resource status and scheduling module, used to calculate the overall resource status of the outpatient clinic, control resource allocation, and schedule patient treatment processes based on resource balance;

[0040] AI control and simulation output module, used to execute outpatient management processes, optimize objective functions, and generate management data and patient feedback.

[0041] The beneficial effects of the present invention: This invention uses generative AI to semantically understand patient symptoms and predict triage, thus achieving intelligent appointment processing. Patients no longer need to manually make tedious selections; the system automatically generates department recommendations, significantly shortening appointment times and improving the convenience and efficiency of patient appointments. The constructed patient priority assessment model and department resource status model make the calculation of appointment times more scientific and reasonable, coordinate the scheduling of appointments and initial consultations, effectively avoid appointment conflicts and long wait times, and improve the overall efficiency and accuracy of outpatient appointments.

[0042] The dynamic triage strategy combines the severity of the patient's condition with the department's real-time resource load, enabling dynamic adjustments in triage. This avoids the limitations of traditional triage, enabling more appropriate placement for both mild and severe patients, improving the efficiency of medical resource utilization and ensuring patient safety. The establishment of an emergency response mechanism further enhances the outpatient department's ability to respond to emergencies.

[0043] Intelligent scheduling of medical resources enables dynamic resource allocation through comprehensive modeling and optimization of resources such as doctors and examination equipment. This accurately matches real-time patient traffic, reduces idle equipment and patient queues, improves resource utilization, and optimizes the overall operational efficiency of the outpatient clinic.

[0044] In terms of assisting and documenting the diagnosis and treatment process, generative AI assists doctors in generating diagnostic ideas and treatment plans, reducing their workload and improving diagnosis and treatment efficiency. Automatically generating structured medical records and follow-up plans makes medical record keeping more standardized and efficient, helping to improve the quality and continuity of medical services.

[0045] The data feedback and optimization mechanism continuously optimizes management models and strategies through in-depth analysis of outpatient operation data. This enables the outpatient management system to continuously adapt to changes in actual needs, achieving continuous improvement and optimization of outpatient management and providing strong support for the development of smart healthcare.

[0046] The system's various modules work together to achieve intelligent management across the entire process, from patient information acquisition to diagnosis and treatment assistance, resource scheduling, and data optimization. The in-depth application of generative AI has given the outpatient management system greater learning and adaptability, enabling it to better meet the needs of diverse patients and complex outpatient management scenarios, thereby improving the overall service level and management efficiency of the outpatient clinic. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0048] Figure 1 This is a flow chart of an outpatient management method based on generative AI of the present invention;

[0049] Figure 2 This is a system block diagram of an outpatient management system based on generative AI in the present invention. DETAILED DESCRIPTION

[0050] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0051] See also Figure 1-Figure 2 As shown, the present invention is an outpatient management method based on generative AI, comprising:

[0052] When a patient visits a doctor through the AI ​​outpatient management system, the system obtains the patient's basic information, symptom description, and historical medical records. A generative AI model is then used to semantically understand the patient's symptoms and conduct preliminary triage predictions, generating possible disease types and corresponding department recommendations.

[0053] Set the target appointment cycle duration, obtain the current appointment volume of each department, doctor's consultation status and expected treatment time, implement the time coordination strategy, combine patient priority and department resources, calculate the optimal appointment time for each patient, and optimize the overall appointment cycle; the implementation of the time coordination strategy also includes:

[0054] Limiting the appointment or initial consultation stage of each patient at any time;

[0055] Calculate the appointment duration of the i-th patient T(i) = T1(i) + T2(i), where T1(i) is the appointment phase duration and T2(i) is the initial consultation waiting phase duration;

[0056] Get the target cycle time T target , calculate the duration objective function

[0057] Establish a patient condition severity assessment model, input patient symptoms, vital signs and other data to generate triage priorities, implement resource balancing strategies based on the real-time number of patients in each department and the doctor's consultation progress, and dynamically adjust the patient triage department and sequence;

[0058] Obtain the doctor's professional field, diagnosis and treatment efficiency and current workload, obtain the use status, maintenance cycle and appointment status of the examination equipment, implement resource modeling optimization strategy, calculate the resource requirements of each link in the diagnosis and treatment process, and optimize the resource allocation objective function; the implementation of resource modeling optimization strategy includes:

[0059] Calculate the resource requirements of department j:

[0060] Doctor resource requirements Among them, P 1,j is the average diagnosis and treatment efficiency of doctor j in department, ) is the time when department j starts to treat patients;

[0061] Check device resource requirements Among them, P 2,j The maximum power used by the device. The time when the device starts to be used, T 2,j The life cycle of the equipment;

[0062] Waiting space resource demand E 3,j (t)=-P 3,j ×T 3,j , where P 3,j is the waiting space unit release rate, T 3,j The average waiting time for patients.

[0063] The execution resource modeling optimization strategy also includes:

[0064] Calculate the resource status of department j at any time t: Resource status = Resource release - First resource consumption, where First resource consumption is the sum of doctor resources and equipment resources consumed during the diagnosis and treatment phase, and Resource release is the sum of waiting space resources released during the patient departure phase.

[0065] Calculate the rate of change of resource status within the time step Δt, and define the resource allocation objective function as minimizing the rate of change.

[0066] Generative AI is used to assist doctors in generating differential diagnosis ideas and treatment plan recommendations, record the diagnosis and treatment process in real time and automatically generate structured medical records, and generate follow-up plans and health management recommendations based on the patient's condition;

[0067] Collect outpatient operation data, use generative AI analysis to identify bottleneck problems, optimize appointment scheduling, triage and resource allocation models and strategies, and present the optimization results to medical staff and patients.

[0068] Intelligent appointment scheduling includes:

[0069] Build a patient priority assessment model to generate appointment priority weights based on the urgency of the patient's condition, the early or late appointment time, and special needs;

[0070] Establish a department resource status model, including the number of doctors in each department, the reception capacity, the number of patients with appointments, and the expected consultation time;

[0071] Calculate the appointment time for patient i, i = 1, 2, ..., n:

[0072] Reservation start time t1(i) start =max(t1(i-1) end ,t2(i-1) start -T buffer ), where t1(i-1) end The appointment end time for patient i-1, t2(i-1) start is the time when patient i-1 first visited the hospital, T buffer Provide buffer time between appointment and initial consultation;

[0073] Time t2(i) when the first visit begins start =max(t1(i) end ,t2(i-1) end ), where t1(i) end The appointment end time for patient i, t2(i-1) end The time when the initial visit of patient i-1 ends.

[0074] An outpatient management system based on generative AI, including:

[0075] Patient information modeling module, used to input and extract patients' basic information, symptom descriptions, historical medical records, vital signs and other data;

[0076] Intelligent appointment scheduling module, used to implement appointment process control, including patient priority assessment, department resource status monitoring and appointment time calculation;

[0077] Dynamic triage module, used to calculate triage priority and dynamically adjust the triage order based on the patient's condition and department resources;

[0078] Resource coordination module, used to coordinate resource allocation among departments and patient treatment order according to target cycle duration;

[0079] A diagnosis and treatment assistance module is used to calculate the resource requirements of each department, generate diagnosis and treatment recommendations, and automatically record medical records;

[0080] Resource status and scheduling module, used to calculate the overall resource status of the outpatient clinic, control resource allocation, and schedule patient treatment processes based on resource balance;

[0081] AI control and simulation output module, used to execute outpatient management processes, optimize objective functions, and generate management data and patient feedback.

[0082] Specifically include the following embodiments:

[0083] Reference Figure 1 , a generative AI-based outpatient management method, comprising: when a patient visits a doctor through the AI ​​outpatient management system:

[0084] Patients enter basic information such as name, age, and gender, as well as a description of symptoms like headache, fever, and cough, through the system client. The system then automatically retrieves the patient's historical medical records. A generative AI model semantically understands the patient's symptom description, analyzing key information such as the nature, duration, and accompanying symptoms. Combined with historical medical records, it generates possible disease types and corresponding department recommendations. For example, a cold might be recommended for consultation with a respiratory specialist.

[0085] The target appointment cycle is set to 3 days, meaning patients are expected to complete their initial consultation within 3 days. The system obtains the current number of appointments for each department, such as 50 appointments for the Department of Respiratory Medicine and 30 appointments for the Department of Gastroenterology. It also obtains the doctor's availability, such as 3 doctors are currently on duty in the Department of Respiratory Medicine, and each doctor is expected to see 20 patients per day. It also obtains the estimated consultation time, such as 15 minutes for each patient with the common cold.

[0086] A patient priority assessment model was constructed, which comprehensively considers the urgency of the patient's condition (e.g., a patient with a fever of 39°C is prioritized over a patient with a common cough), the early or late appointment time, and whether there are special needs (e.g., pregnant women, the elderly), to generate an appointment priority weight for each patient. A department resource status model was also established, including the number of physicians in each department, each physician's capacity, the number of currently scheduled patients, and their estimated appointment times.

[0087] Calculate the appointment time for patient i, i = 1, 2, ..., n:

[0088] Coordination appointment stage: t1(i) start =max(t1(i-1) end ,t2(i-1) start -T buffer ), where T buffer Set to 1 day to ensure a reasonable buffer time between appointments and initial visits. For example, if patient i-1's appointment ends at 10:00 AM on day 1 and patient i-1's initial visit starts at 8:00 AM on day 2, then patient i's appointment start time must be at least 10:00 AM on day 1.

[0089] Coordination of initial diagnosis stage: t2(i) start =max(t1(i) end ,t2(i-1) end ) to ensure that the first consultation starts after the appointment ends and no earlier than the end time of the previous patient's first consultation.

[0090] In the process of coordinating patient appointments and initial consultations, it is necessary to limit the appointment or initial consultation stage of each patient at any time to avoid resource conflicts. Calculate the appointment duration of the i-th patient T(i) = T1(i) + T2(i), obtain the target cycle duration of 3 days, and calculate the duration objective function By optimizing the algorithm, the appointment time and duration of each patient are adjusted to make the overall appointment cycle as close to the target cycle as possible, thereby improving appointment efficiency.

[0091] A patient severity assessment model is established, inputting patient symptoms and vital signs data such as temperature, blood pressure, and heart rate to generate triage priorities. For example, patients with high fever and dyspnea are prioritized as Level 1, while those with the common cold are prioritized as Level 3. A resource balancing strategy is implemented based on the real-time number of patients in each department and the doctor's availability, dynamically adjusting the department and order of patient triage. If the respiratory department has a long waiting list, some patients with milder symptoms can be transferred to other departments or scheduled for a later appointment.

[0092] Calculate the resource requirements of department j during the diagnosis and treatment process:

[0093] Calculating physician resource requirements For example, the average diagnosis and treatment power of respiratory physicians P 1,j 4 people per hour, the time to start treating patients It is 8:00 am on the 3rd day, so at 9:00 am on the 3rd day, the doctor resource demand is 4×(9-8)=4 people.

[0094] Calculate and check device resource requirements Assume that the maximum power P of a certain examination equipment in the Department of Respiratory Medicine is 2,j 10 times per hour, starting time It is 8:00 am on the third day, and the usage cycle duration is T 2,j For 1 hour, then at 8:30 am on the third day, the equipment resource requirements are

[0095] Calculate waiting space resource requirements E 3,j (t)=-P 3,j ×T 3,j , the unit release rate P of the waiting space of the respiratory department 3,j 5 people per hour, the average waiting time for patients is T 3,j If the waiting time is 0.5 hours, then the waiting space resource demand is -5×0.5=-2.5, which means that 2.5 waiting spaces are released every hour.

[0096] For department j, physician resources and examination equipment resources are consumed during the diagnosis and treatment phase, and waiting room resources are released when the patient leaves the department. The resource state of department j at any time t is calculated as = resource release - first resource consumption. The rate of change of the resource state within a time step Δt = 1 hour is calculated. The resource allocation objective function is to minimize this rate of change, ensuring that the department's resource state is as stable as possible and avoiding excessive resource fluctuations.

[0097] Generative AI assists doctors in generating differential diagnosis ideas and treatment recommendations. For example, for patients with fever, AI generates possible differential diagnoses, including colds, pneumonia, and influenza, and provides corresponding examination and treatment recommendations. Key information from the diagnosis and treatment process, such as diagnosis results and medication usage, is recorded in real time, automatically generating structured medical records. Follow-up plans and health management recommendations are generated based on the patient's condition and treatment plan, such as recommending a blood test be repeated one week later.

[0098] Collect various data during outpatient operations, such as patient waiting time, physician workload, and equipment utilization. Use generative AI to analyze data and identify bottlenecks in outpatient management, such as the long patient waiting time between 10:00 and 11:00 a.m. in the respiratory department. Based on the analysis results, optimize appointment scheduling, triage, and resource allocation models and strategies. For example, when making the next appointment, appropriately reduce the number of appointments for that time period or increase the number of temporary physician visits. Control the AI ​​outpatient management system to display the optimized outpatient management results to medical staff and patients, such as displaying adjusted appointment times and triage arrangements on the system interface.

[0099] Example 2

[0100] Reference Figure 2 , an outpatient management system based on generative AI, including:

[0101] Patient information modeling module: provides a user interface for patients to enter basic information, symptom descriptions, etc., and connects with the hospital's electronic medical record system to extract patient historical medical records, vital signs and other data.

[0102] Intelligent appointment scheduling module: includes a patient priority evaluation submodule, which calculates the patient appointment priority based on the set evaluation model; a department resource status monitoring submodule, which obtains the number of appointments in each department, doctor's consultation status, etc. in real time; and an appointment time calculation submodule, which calculates the patient appointment time based on the time coordination strategy.

[0103] Dynamic triage module: includes a condition assessment submodule that uses an assessment model to generate patient triage priorities; and a resource balancing submodule that dynamically adjusts the triage order based on the department's real-time resource load.

[0104] Resource coordination module: Receive target cycle time, coordinate resource allocation among departments and patient visit sequence to ensure smooth outpatient process.

[0105] Diagnosis and treatment assistance module: resource demand calculation submodule, which calculates the resource demand in the diagnosis and treatment process of each department; diagnosis and treatment recommendation generation submodule, which uses generative AI to generate diagnostic ideas and treatment plans; medical record recording submodule, which automatically generates structured medical records and follow-up plans.

[0106] Flywheel energy storage and scheduling module (outpatient scenario conversion): resource status calculation submodule, calculates the overall resource status of the outpatient clinic; resource scheduling submodule, schedules the patient treatment process based on resource balance to ensure the rational use of resources.

[0107] AI control and simulation output module: executes outpatient management processes and uses optimization algorithms to dynamically optimize cycle and resource allocation objective functions; data display sub-module generates management data reports and displays outpatient management results and feedback information to medical staff and patients.

[0108] Throughout the specification, references to terms such as "one embodiment," "example," or "specific example" indicate that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0109] The above content is merely an example and explanation of the concept of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the claims, they should all fall within the scope of protection of the present invention.

Claims

1. A generative AI-based outpatient management method, characterized by: include: When a patient visits a doctor through the AI ​​outpatient management system, the system obtains the patient's basic information, symptom description, and historical medical records. A generative AI model is then used to semantically understand the patient's symptoms and conduct preliminary triage predictions, generating possible disease types and corresponding department recommendations. Set the target appointment cycle duration, obtain the current appointment volume of each department, doctor's consultation status and expected treatment time, implement time coordination strategy, combine patient priority and department resources, calculate the optimal appointment time for each patient, and optimize the overall appointment cycle; Establish a patient condition severity assessment model, input patient symptoms, vital signs and other data to generate triage priorities, implement resource balancing strategies based on the real-time number of patients in each department and the doctor's consultation progress, and dynamically adjust the patient triage department and sequence; Obtain the doctor's professional field, diagnosis and treatment efficiency and current workload, obtain the use status, maintenance cycle and appointment status of the examination equipment, implement resource modeling optimization strategy, calculate the resource requirements of each link in the diagnosis and treatment process, and optimize the resource allocation objective function; Generative AI is used to assist doctors in generating differential diagnosis ideas and treatment plan recommendations, record the diagnosis and treatment process in real time and automatically generate structured medical records, and generate follow-up plans and health management recommendations based on the patient's condition; Collect outpatient operation data, use generative AI analysis to identify bottleneck problems, optimize appointment scheduling, triage and resource allocation models and strategies, and present the optimization results to medical staff and patients.

2. The outpatient management method based on generative AI according to claim 1, characterized in that: Intelligent appointment scheduling includes: Build a patient priority assessment model to generate appointment priority weights based on the urgency of the patient's condition, the early or late appointment time, and special needs; Establish a department resource status model, including the number of doctors in each department, the reception capacity, the number of patients with appointments, and the expected consultation time; Calculate the appointment time for patient i, i = 1, 2, ..., n: Reservation start time t1(i) start =max(t1(i-1) end ,t2(i-1) start -T buffer ), where t1(i-1) end The appointment end time for patient i-1, t2(i-1) start is the time when patient i-1 first visited the hospital, T buffer Provide buffer time between appointment and initial consultation; Time t2(i) when the first visit begins start =max(t1(i) end ,t2(i-1) end ), where t1(i) end The appointment end time for patient i, t2(i-1) end The time when the initial visit of patient i-1 ends.

3. The outpatient management method based on generative AI according to claim 1, characterized in that: The execution time coordination strategy also includes: Limiting the appointment or initial consultation stage of each patient at any time; Calculate the appointment duration of the i-th patient T(i) = T1(i) + T2(i), where T1(i) is the appointment phase duration and T2(i) is the initial consultation waiting phase duration; Get the target cycle time T target , calculate the duration objective function 4. The outpatient management method based on generative AI according to claim 1, characterized in that: The execution resource modeling optimization strategy includes: Calculate the resource requirements of department j: Doctor resource requirements Among them, P 1,j is the average diagnosis and treatment efficiency of doctor j in department, ) is the time when department j starts to treat patients; Check device resource requirements Among them, P 2,j The maximum power used by the device. The time when the device starts to be used, T 2,j The life cycle of the equipment; Waiting space resource demand E 3,j (t)=-P 3,j ×T 3,j , where P 3,j is the waiting space unit release rate, T 3,j The average waiting time for patients.

5. The outpatient management method based on generative AI according to claim 1, characterized in that: The execution resource modeling optimization strategy also includes: Calculate the resource status of department j at any time t: Resource status = Resource release - First resource consumption, where First resource consumption is the sum of doctor resources and equipment resources consumed during the diagnosis and treatment phase, and Resource release is the sum of waiting space resources released during the patient departure phase. Calculate the rate of change of resource status within the time step Δt, and define the resource allocation objective function as minimizing the rate of change.

6. A generative AI-based outpatient management system, applied to the outpatient management method according to any one of claims 1 to 5, characterized in that: include: Patient information modeling module, used to input and extract patients' basic information, symptom descriptions, historical medical records, vital signs and other data; Intelligent appointment scheduling module, used to implement appointment process control, including patient priority assessment, department resource status monitoring and appointment time calculation; Dynamic triage module, used to calculate triage priority and dynamically adjust the triage order based on the patient's condition and department resources; Resource coordination module, used to coordinate resource allocation among departments and patient treatment order according to target cycle duration; A diagnosis and treatment assistance module is used to calculate the resource requirements of each department, generate diagnosis and treatment recommendations, and automatically record medical records; Resource status and scheduling module, used to calculate the overall resource status of the outpatient clinic, control resource allocation, and schedule patient treatment processes based on resource balance; AI control and simulation output module, used to execute outpatient management processes, optimize objective functions, and generate management data and patient feedback.

Images