Doctor resource allocation method and device, electronic equipment and readable storage medium

Abstract

The invention provides a doctor resource allocation method and device, electronic equipment and a readable storage medium, and relates to the technical field of computers. In the method, a doctor-seeing record of a patient is obtained, the doctor-seeing record comprises a doctor diagnosis score and a doctor, and the doctor diagnosis score is obtained by scoring according to the disease degree of the patient when the doctor sees a doctor; according to the doctor-seeing record, predicting the doctor-seeing node by using a preset target prediction model to obtain a predicted doctor-seeing node; the patients are distributed to a doctor-seeing group where the doctor is located, the professional level, the working state and the bidding score of each member in the doctor-seeing group are obtained, and the members in the doctor-seeing group have the authority to check the doctor-seeing records of the patients; the bidding score is dynamically adjusted according to the professional level, the predicted doctor seeing node, the doctor diagnosis score and the working state, the fitness bid is obtained, the member matched with the maximum fitness bid is selected and pushed to the patient, and the medical service efficiency can be improved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and specifically to a method, apparatus, electronic device, and readable storage medium for allocating doctor resources. Background Technology

[0002] In healthcare, information technology is being used to optimize the allocation of medical resources and improve the efficiency of medical services. With the development of internet technology, online matching services between doctors and patient needs have become an important direction for improving the healthcare experience. One related technology, to ensure accurate matching of doctors and patient needs, uses historical medical records for doctor allocation. While these records facilitate patient tracking, multiple patients may be assigned to the same doctor. However, during peak demand periods, this method can lead to wasted resources for doctors who haven't participated in historical consultations, and overburden those who have, thus impacting the efficiency of medical services. Summary of the Invention

[0003] This application provides a method, apparatus, electronic device, and readable storage medium for allocating physician resources, which can improve the efficiency of medical services.

[0004] The technical solution of this application embodiment is as follows: In a first aspect, embodiments of this application provide a method for allocating physician resources, the method comprising: Obtain the patient's medical records, which include the doctor's diagnostic score and the attending physician. The doctor's diagnostic score is obtained by the doctor during the consultation based on the severity of the patient's condition. Based on the medical records, a preset target prediction model is used to predict the medical visit node, resulting in a predicted medical visit node. The patient is assigned to the medical visit group of the attending physician, and the professional level, work status, and bidding score of each member in the medical visit group are obtained. Each member in the medical visit group has the right to view the patient's medical records. The bidding score is calculated by the members in the medical visit group using a preset bidding function based on the professional level, the physician's diagnostic score, the work status, and the predicted medical visit node. The bidding score is dynamically adjusted using the professional level, the predicted visit node, the doctor's diagnosis score, and the work status to obtain the suitability bid. Members matching the maximum suitability bid are then selected and pushed to the patient.

[0005] In the above technical solution, the patient's medical records are first obtained. These records include the doctor's diagnostic score and the attending physician's information. The doctor's diagnostic score is obtained by the doctor based on the severity of the patient's condition during the consultation. The information recorded in the medical records provides data support for subsequent calculations. Based on the medical records, a pre-set target prediction model is used to predict the consultation node, resulting in a predicted consultation node. This prediction allows for the matching of appropriate members to the predicted consultation node, improving service efficiency. The patient is then assigned to the consultation group of the attending physician. Members in the physician's group have high professional and job relevance, providing support for subsequent resource allocation. The professional level, work status, and bidding score of each member in the consultation group are obtained. All members in a group have access to view a patient's medical records, allowing other members in the group to understand the patient's condition. This alleviates the patient tracking pressure on attending physicians and avoids overburdening them. The bidding score is calculated by members in the group using a preset bidding function based on their professional level, doctor's diagnostic score, work status, and predicted appointment time. Members submit their bidding scores to compete for patients, avoiding the waste of doctor resources. The bidding score is dynamically adjusted using professional level, predicted appointment time, doctor's diagnostic score, and work status to obtain a suitable bid. Members with the highest suitable bid are selected and pushed to the patient. Through dynamic adjustment, more suitable members can be matched to patients to provide services, improving the efficiency of medical services.

[0006] In some embodiments of this application, the step of dynamically adjusting the bidding score using the professional level, the predicted consultation node, the doctor's diagnostic score, and the work status to obtain a suitable bid includes: If the doctor's diagnostic score is greater than a preset scoring threshold, the professional level and the attending physician are quantified to obtain the level quantified score corresponding to the professional level and the consultation quantified score corresponding to the attending physician; the level quantified score and the consultation quantified score are summed, and the sum is used as the bidding reward score; The bidding penalty score is calculated based on the working status and the predicted medical visit node, and the sign of the bidding penalty score is set to a negative number. The bidding reward score and the bidding penalty score are subjected to a nonlinear game to obtain the game result. Obtain the first weight corresponding to the bidding score, the second weight corresponding to the bidding reward score, and the third weight corresponding to the bidding penalty score, wherein the sum of the first weight, the second weight, and the third weight is 1; The second weight and the third weight are adjusted using the game results. The results of multiplying the first weight by the bidding score, the adjusted second weight by the bidding reward score, and the adjusted third weight by the bidding penalty score are summed to obtain the fit bid.

[0007] In some embodiments of this application, the step of performing nonlinear game processing on the bidding reward score and the bidding penalty score to obtain the game result includes: The bidding reward score and the bidding penalty score are subjected to nonlinear game processing using a preset game formula to obtain the game result; The game theory formula is expressed as follows: Where R represents the bidding reward score, P represents the bidding penalty score, and λ, μ, and c are adjustment parameters. represents the historical average fitness value, ni represents the number of times it was selected in history, t represents the total number of bids in history, and V represents the game outcome.

[0008] In some embodiments of this application, the work status includes idle time, number of patients received, and proportion of urgent tasks; the calculation of the bidding penalty score based on the work status and the predicted visit node includes: The ratio of the number of patients received to a preset threshold for the number of patients received is calculated to obtain the saturation of the number of patients received. The saturation of the number of patients received is then weighted and summed with the proportion of emergency tasks to obtain the load index. The processing time for the number of patients is estimated to obtain the task processing time. The processing time for the medical items in the predicted medical node is predicted to obtain the new task time. The task processing time and the new task time are summed. The sum is compared with the idle time to obtain the predicted time pressure value. The time matching degree is obtained by matching the appointment time in the predicted appointment node with the idle time; The load index, the predicted time pressure, and the time matching degree are penalized to obtain the bidding penalty score.

[0009] In some embodiments of this application, the step of calculating a penalty score for the load index, the predicted time pressure value, and the time matching degree to obtain a bidding penalty score includes: The load index, the predicted time pressure, and the time matching degree are penalized using a preset bidding penalty formula to obtain a bidding penalty score. The bidding penalty formula is expressed as follows: Penalty=δ*TWM+(1-δ)*max(1,α*RWI+β*TPF(t)) Wherein, Penalty represents the bidding penalty score, TWM represents the time matching degree, RWI represents the load index, TPF(t) represents the predicted time pressure value, max() represents the maximization function, and δ, α, and β all represent adjustment parameters.

[0010] In some embodiments of this application, adjusting the second weight and the third weight using the game result includes: If the game result is greater than a preset adjustment threshold, the second weight is increased and the third weight is decreased, while the sum of the second weight and the third weight remains unchanged. If the game result is less than a preset adjustment threshold, the second weight is reduced, the third weight is increased, and the sum of the second weight and the third weight remains unchanged. If the game result is equal to the preset adjustment threshold, the second weight and the third weight are set to the same value.

[0011] In some embodiments of this application, the target prediction model is obtained in the following manner: Obtain multiple historical medical records, each of which includes multiple historical medical items, the historical medical time corresponding to each historical medical item, the next medical item corresponding to the historical medical item, and the next medical time corresponding to the next medical item. Multiple medical visits and multiple historical medical visits are input into a preset initial prediction model in a preset batch to obtain predicted medical visits and predicted medical visits. Calculate the values ​​of the loss functions for the predicted medical visit items, the predicted medical visit time, the next node medical visit items, and the next node medical visit time. Use the values ​​of the loss functions to train the initial prediction model in reverse until the preset training conditions are met, and obtain the target prediction model.

[0012] Secondly, embodiments of this application provide a doctor resource allocation device, the device comprising: The data acquisition module is used to acquire the patient's medical records, which include the doctor's diagnostic score and the attending physician. The doctor's diagnostic score is obtained by the doctor scoring the patient's condition during the consultation. The data prediction module is used to predict the visit node based on the medical records using a preset prediction model, and obtain the predicted visit node. The data grouping module is used to assign the patient to the consultation group of the attending physician, and to obtain the professional level, work status and bidding score of each member in the consultation group. The members in the consultation group have the right to view the patient's medical records. The bidding score is calculated by the members in the consultation group using a preset bidding function based on the professional level, the doctor's diagnosis score, the work status and the predicted consultation node. The data matching module is used to dynamically adjust the bidding score using the professional level, the predicted visit node, the doctor's diagnosis score, and the work status to obtain the suitability bid, and select members that match the maximum suitability bid to push to the patient.

[0013] Thirdly, embodiments of this application provide an electronic device including a processor, a memory, a user interface, a communication bus, and a network interface. The processor, the memory, the user interface, and the network interface are respectively connected to the communication bus. The memory is used to store instructions. The user interface and the network interface are used to communicate with other devices. The processor is used to execute the instructions stored in the memory to cause the electronic device to perform the method described in any one of the first aspects.

[0014] Fourthly, embodiments of this application provide a computer-readable storage medium storing instructions that, when executed, perform the method described in any one of the methods provided in the first aspect above.

[0015] In summary, one or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: 1. The system first obtains the patient's medical records, including the doctor's diagnostic score and the attending physician's information. The doctor's diagnostic score is obtained by the doctor based on the severity of the patient's condition during the consultation. The information recorded in the medical records provides data support for subsequent calculations. Based on the medical records, a pre-set target prediction model is used to predict the consultation node, resulting in a predicted consultation node. This prediction allows for the matching of appropriate members to the predicted consultation node, improving service efficiency. Patients are assigned to the consultation group of the attending physician. Members in the physician's group have high professional and job relevance, supporting subsequent resource allocation. The system obtains the professional level, work status, and bidding score of each member in the consultation group. The consultation score... All members of the group have access to view patients' medical records, allowing other members in the group to understand the patient's condition. This alleviates the patient tracking pressure on attending physicians and avoids overload. The bidding score is calculated by members in the group using a preset bidding function based on their professional level, doctor's diagnostic score, work status, and predicted appointment time. Members submit their bidding scores to compete for patients, avoiding waste of doctor resources. The bidding score is dynamically adjusted using professional level, predicted appointment time, doctor's diagnostic score, and work status to obtain a suitable bid. Members with the highest suitable bid are selected and pushed to patients. Through dynamic adjustment, more suitable members can be matched to patients, improving the efficiency of medical services. Therefore, this effectively solves the problem in related technologies where, during peak demand periods, the above method can easily lead to wasted doctor resources for doctors who have not participated in previous appointments, and can easily lead to overload for doctors who have participated in previous appointments, thus affecting the efficiency of medical services.

[0016] 2. By calculating bidding reward points and bidding penalty points, and adjusting the bidding score using these points, the resulting appropriate bid can meet the patient's needs, thereby improving service quality and efficiency.

[0017] 3. By applying non-linear game theory to the bidding reward points and bidding penalty points, the competitiveness of eligible members can be improved, achieving maximum fit and thus improving the efficiency of medical services. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating a doctor resource allocation method provided in one embodiment of this application; Figure 2 yes Figure 1 A flowchart illustrating a sub-step of step S400; Figure 3 yes Figure 2 A flowchart illustrating a sub-step of step S430 in the middle section; Figure 4 yes Figure 2 A flowchart illustrating a sub-step of step S460; Figure 5 This is a schematic diagram of the structure of a doctor resource allocation device provided in one embodiment of this application; Figure 6 This is a schematic diagram of the structure of an electronic device provided in one embodiment of this application. Detailed Implementation

[0019] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0020] In the description of the embodiments of this application, the words "for example" or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design that is described as "for example" or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design options. Rather, the use of the words "for example" or "for instance" is intended to present the relevant concepts in a specific manner.

[0021] In the description of the embodiments of this application, the term "multiple" means two or more. For example, multiple systems means two or more systems, and multiple screen terminals means two or more screen terminals. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature. The terms "comprising," "including," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0022] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards of the relevant countries and regions. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through prompts, pop-ups, or redirection to a confirmation page. Only after obtaining the user's separate permission or consent is the necessary user-related data for the proper functioning of the embodiments of this application obtained.

[0023] This application provides a method, apparatus, electronic device, and readable storage medium for allocating doctor resources. The method first acquires a patient's medical records, including a doctor's diagnostic score and the attending physician's information. The doctor's diagnostic score is obtained by the doctor based on the severity of the patient's condition during the consultation, and the information recorded in the medical records provides data support for subsequent calculations. Based on the medical records, a preset target prediction model is used to predict the consultation node, resulting in a predicted consultation node. This prediction allows for the matching of appropriate members to the predicted consultation node, improving service efficiency. The patient is then assigned to a consultation group belonging to the attending physician. Members of the physician's group have high professional and job relevance, providing support for subsequent resource allocation. The method also obtains the professional information of each member in the consultation group. The system considers factors such as professional level, work status, and bidding score. Members in the patient consultation group have access to view patients' medical records, allowing other members in the group to understand the patient's condition. This alleviates the patient tracking pressure on attending physicians and avoids overburdening them. The bidding score is calculated by members in the consultation group using a preset bidding function based on their professional level, physician diagnosis score, work status, and predicted consultation time. Members bid for patients, avoiding waste of physician resources. The bidding score is dynamically adjusted based on professional level, predicted consultation time, physician diagnosis score, and work status to obtain a suitable bid. Members with the highest suitable bid are selected and recommended to patients. This dynamic adjustment matches patients with more suitable members, improving the efficiency of medical services.

[0024] It should be noted that this physician resource allocation method is used for hospital department and ward management to achieve a balanced distribution of physician resources and improve the quality and efficiency of medical services. It can also be applied to emergency and intensive care units, enabling rapid response and efficient treatment, thus improving the efficiency of medical services. Furthermore, it can be applied to medical collaboration and tiered medical services to achieve teamwork, rationally allocate medical resources, and improve service efficiency.

[0025] The technical solutions provided in the embodiments of this application will be further described below with reference to the accompanying drawings.

[0026] Reference Figure 1 , Figure 1 This is a schematic flowchart of the doctor resource allocation method provided in this application embodiment. The doctor resource allocation method is applied to a doctor resource allocation device and is executed by a processor in an electronic device or a readable storage medium. The doctor resource allocation method includes steps S100, S200, S300, and S400.

[0027] Step S100: Obtain the patient's medical records, which include the doctor's diagnostic score and the attending physician. The doctor's diagnostic score is obtained by the doctor scoring the patient's condition during the consultation.

[0028] In one embodiment, a patient's medical record includes a doctor's diagnostic score and the attending physician, as well as the patient's medical procedures, the corresponding consultation time, and a summary of their condition. During routine consultations, each time a doctor diagnoses a patient, they input the patient's medical record information into the computer, write a summary of their condition, and score and record the severity of the patient's condition. This information is stored using a unique identifier, and a record of each patient's medical record is maintained. When viewing the record, a preset read function is used to retrieve the patient's medical record, providing data support for subsequent calculations. The preset read function can be either `read()` or `open()`. The doctor's diagnostic score is represented by a value from 1 to 10. A higher score indicates a more severe condition; a score greater than 6 indicates a critically ill patient who needs to be referred to a physician at the attending physician level or higher.

[0029] Step S200: Based on the medical records, the medical visit node is predicted using a preset target prediction model to obtain the predicted medical visit node.

[0030] In one embodiment, the medical record includes the medical items and the corresponding medical time. The medical items and medical time are input into a preset target prediction model. The target prediction model predicts the next medical node to obtain a predicted medical node. The predicted medical node includes the predicted medical items and the predicted medical time, so that the corresponding members can be matched according to the predicted medical node in the future to improve service efficiency.

[0031] Specifically, the target prediction model is obtained as follows: Multiple historical medical records are acquired, each including multiple historical medical items, the corresponding historical medical time, the next-node medical item, and the next-node medical time; the multiple medical items and multiple historical medical times are input into a preset initial prediction model in a preset batch to obtain the predicted medical items and predicted medical times; the loss function values ​​for the predicted medical items, predicted medical times, next-node medical items, and next-node medical times are calculated, and the initial prediction model is trained in reverse using the loss function values ​​until the preset training conditions are met, thus obtaining the target prediction model.

[0032] In some possible embodiments of this application, historical medical records include multiple historical medical items, the historical medical time corresponding to each historical medical item, the next-node medical item corresponding to each historical medical item, and the next-node medical time corresponding to the next-node medical item. During routine consultations, each time a doctor diagnoses a patient, they enter the patient's historical medical record information into the computer, write a summary of the patient's condition, and score and record the severity of the patient's condition. This information is stored using a unique identifier, recording the historical medical record for each patient. A next-step consultation suggestion is also provided, which includes the next-node medical item corresponding to the historical medical item and the next-node medical time corresponding to the next-node medical item. By conducting consultations with multiple patients over a long period, a large amount of historical medical record data can be recorded. When reviewing this data, multiple historical medical records for multiple patients can be directly retrieved using a preset read function. The content recorded in these historical medical records can then be used to provide training data for subsequent model training.

[0033] The preset batch size is a hyperparameter of the model, which can be adjusted and can take values ​​such as 1, 2, and 4. The preset initial prediction model is a deep neural network model, which can be a recurrent neural network model or a variant thereof. Multiple medical visits and multiple historical medical visits are input into the preset initial prediction model according to the preset batch size to obtain predicted medical visits and predicted medical visits and times. For example, the preset batch size is set to 1, and only one historical medical visit and one historical medical visit and time are input at a time in each iteration of training. Then, the historical medical visits and times are transformed into features. This process can be handled using the embedding method to convert them into feature vectors corresponding to the historical medical visits and time features corresponding to the historical medical visits and times. The feature vectors of the medical visits and the time features are then fused by vector concatenation to obtain the input features. The initial prediction model is used to perform feature extraction, nonlinear processing, and normalization on the input features to output predicted medical visits and predicted medical visits and times.

[0034] The logarithmic loss function is then used to calculate a first loss function between the predicted medical item and the next-node medical item, and a second loss function between the predicted medical time and the next-node medical time. The first and second loss functions are then weighted and summed to obtain the value of the loss function. Specifically, the sum of the weights of the second and first loss functions is set to 1, and the weight of the second loss function is greater than the weight of the first loss function, with the difference between their weights not exceeding 0.5. Based on the calculated loss function value, backpropagation is performed on the initial prediction model to update its parameters until the preset training conditions are met, resulting in the target prediction model. This target prediction model exhibits superior performance, accurately predicting both the next-node medical item and the next-node medical time, providing data support for subsequent resource allocation.

[0035] The preset training conditions can be: the loss function value tends to stabilize, and training ends after stabilization; or a set number of training iterations is set, and training ends after the set number of iterations is reached. Alternatively, a combination of the loss function stabilization and a set number of training iterations can be used, ending training upon reaching either of these conditions, thus obtaining the target prediction model.

[0036] Step S300: Assign the patient to the consultation group of the attending physician, and obtain the professional level, work status and bidding score of each member in the consultation group. Members in the consultation group have the right to view the patient's medical records. The bidding score is calculated by the members in the consultation group using a preset bidding function based on the professional level, doctor's diagnosis score, work status and predicted consultation node.

[0037] In one embodiment, in actual work, the attending physician has a corresponding department, which is used as the treatment group. Patients can also be assigned to the treatment group of the attending physician based on the attending physician's department or the patient's registered department. Because they belong to the same department, the members of the attending physician's group have a high degree of professional and work relevance, and nurses in the department can also follow up with ordinary patients under the doctor's guidance, reducing the doctor's patient tracking burden. In each treatment group, the professional level, work status, and bidding score of each member are recorded, and then a preset read function is used to retrieve the records or the professional level, work status, and bidding score reported by the members. The preset read function can be either the `read()` function or the `open()` function. By retrieving the professional level, work status, and bidding score of each member in the treatment group, members can report their own relevant information. This reported personal information is more closely aligned with the member's actual work. Patient allocation competition based on reported bidding scores replaces the original system's automatic patient allocation, avoiding excessive workload for doctors and waste of doctor resources, thus improving service efficiency.

[0038] It should be noted that the professional level reflects the professional competence of each member in the consultation group. This level can be recorded and stored in the hospital's system when members are hired or promoted. The professional levels include chief physician, associate chief physician, attending physician, general physician, and nurse, each corresponding to a different level: chief physician (5), associate chief physician (4), attending physician (3), general physician (2), and nurse (1). The work status includes free time, number of patients seen, and proportion of emergency tasks. This can be reported by members in the consultation group or predicted based on patients' registration records and consultation status. The bidding score is reported by members in the consultation group based on their own situation. The reporting process is done through a reporting window popped up by the hospital. After members fill in the information according to the situation, the information is returned and stored. It should also be noted that the hospital's system updates patients' consultation records in real time. Each time a doctor sees a patient, the system needs to record and store the information to facilitate data retrieval and support the rational allocation of resources in the future, thereby improving service efficiency.

[0039] In this system, all members in the consultation group have the authority to view the patient's medical records, enabling other members in the consultation group to understand the patient's condition. This can alleviate the patient tracking pressure on the attending physician and avoid overburdening the physician. The bidding score is calculated by members in the consultation group using a preset bidding function based on professional level, physician diagnosis score, work status, and predicted consultation node. Members submit their bidding scores to compete for patients, thus avoiding the waste of physician resources.

[0040] In one embodiment, a pre-defined bidding function is used to calculate the bidding score based on professional level, doctor's diagnostic score, work status, and predicted visit time. The formula is as follows: Where Bid represents the bidding score, L represents the professional level, Lmax represents the highest professional level, T represents the predicted consultation node, T1 represents the idle time in the working state, Trange is the time normalized value, S represents the doctor's diagnosis score, and a and b are both adjustment parameters.

[0041] It's important to note that the professional matching ability can be demonstrated by multiplying the ratio of the professional level to the highest professional level with the doctor's diagnostic score. The lower the professional level, the smaller the ratio to the highest professional level, and consequently, the smaller the product of the ratio and the doctor's diagnostic score. When two doctors are competing, the member with the higher professional level will have a higher professional matching ability score and is more likely to win patients. The predicted appointment time and idle time are then calculated and normalized to determine the time matching degree. The predicted appointment time is subtracted from the idle time, and the absolute value is taken to avoid negative numbers in the working time process, which would affect the calculation.

[0042] For example, a = 0.6, b = 0.4, Trange = 1 hour (60 minutes), S = 7, T = 10:00, and there are two doctors, A and B, in the consultation group. Doctor A's professional level is 5, currently busy, with an available time of 10:20. Doctor B's professional level is 4, currently free, with an available time of 9:30, and a maximum professional level of 5. The calculated results of their reported bidding scores are as follows: Doctor A: 0.6*5 / 5*6 + 0.4*(1 - |10:20-10:00| / 60) = 3.6 + 0.4*(1 - 20 / 60) ≈ 3.864 Doctor B: 0.6*4 / 5*6 + 0.4*(1 - |9:30-10:00| / 60) = 2.88 + 0.4*(1 - 30 / 60) = 3.08 By having doctors set their own bids, adjustments can be made based on the bidding scores to ensure a more rational allocation of resources and improve service efficiency.

[0043] Step S400: Dynamically adjust the bidding score based on professional level, predicted consultation time, doctor's diagnosis score, and work status to obtain the suitability bid, and select members that match the maximum suitability bid to push to the patient.

[0044] like Figure 2As shown, the bidding score is dynamically adjusted based on professional level, predicted consultation time, doctor's diagnostic score, and work status to obtain a suitable bid, including but not limited to the following steps: Step S410: If the doctor's diagnostic score is greater than the preset scoring threshold, the professional level and the attending physician are quantified separately to obtain the level quantification score corresponding to the professional level and the consultation quantification score corresponding to the attending physician.

[0045] In one embodiment, the preset scoring threshold can be 6, or other values, determined according to the doctor's diagnostic scoring criteria described above, which will not be elaborated here. If the doctor's diagnostic score exceeds the preset threshold, it indicates that the patient is a critically ill patient and requires a specialist or attending physician to monitor changes in their condition. The professional level and attending physician are quantified separately to obtain the quantified score corresponding to the professional level and the quantified score corresponding to the attending physician. The quantification method is as follows: they are represented as corresponding numerical values. Based on the above, the professional level can be directly quantified, and the quantified value for the attending physician is 1. This quantification prepares for subsequent calculations.

[0046] In another embodiment, if the doctor's diagnostic score is below a preset threshold, it indicates that the patient has a mild condition and does not require a specialist or attending physician to monitor changes in their condition. Instead, the patient can be assigned to a nurse or general practitioner for monitoring. In this case, only the professional level is quantified to obtain a level-quantified score. This allows for the subsequent mapping of professional levels to doctor's diagnostic scores, preventing higher-level professionals from treating low-scoring patients and wasting resources. The level-quantified score is used as a bidding reward; the higher the professional level, the greater the likelihood of successfully competing for a patient.

[0047] Step S420: Sum the grade quantification score and the medical visit quantification score, and use the sum as the bidding reward score.

[0048] In one embodiment, when a doctor's diagnostic score exceeds a preset threshold, the patient is preferentially assigned to a doctor with a better understanding of the patient's condition, provided the doctor's professional level is the same. When professional levels differ, the doctor's role is considered a bonus to enhance their competitiveness. The quantified scores for both professional level and patient visit are summed to further increase competitiveness; this sum is used as a bidding bonus, which can be used to adjust bidding scores and optimize allocation in subsequent transactions.

[0049] Step S430: Calculate the bidding penalty score based on the working status and predicted medical visit node, and set the sign of the bidding penalty score to a negative number.

[0050] In one embodiment, to avoid excessive workload and reduce the competitiveness of employees working long hours, a bidding penalty score is set. This penalty score is calculated based on work status and predicted appointment times, allowing for subsequent adjustments to the bidding score. The bidding penalty score is set to a negative sign so that the bidding score can be reduced during adjustments. In one embodiment, the work status includes idle time, number of patients treated, and the proportion of urgent tasks. For example... Figure 3 As shown, the bidding penalty score is calculated based on the work status and predicted visit time, including but not limited to the following steps: Step S431: Calculate the ratio of the number of patients received to the preset threshold for the number of patients received to obtain the saturation of the number of patients received. Then, sum the saturation of the number of patients received with the proportion of emergency tasks to obtain the load index.

[0051] In some possible embodiments of this application, the number of patients received includes patients who made online appointments and patients with current appointments. The threshold for the number of patients received is the maximum number of patients a doctor can receive on a given day or within a given period. The ratio of the number of patients received to this threshold is used to calculate the patient saturation level, which reflects the doctor's current workload. Emergency tasks may also be added, potentially disrupting the original schedule. The emergency task ratio represents the ratio of currently received emergency tasks to the number of patients received. A weighted sum of the patient saturation level and the emergency task ratio yields a load index, which reflects the current workload for subsequent calculations and helps avoid overwork. It should be noted that the weighted sum of the patient saturation level and the emergency task ratio involves setting a weight for the patient saturation level and a weight for the emergency task ratio. Since emergency tasks are not frequent, the weight for the patient saturation level is set greater than the weight for the emergency task ratio. This allows for adjustment of the load index even when emergency tasks are present, accurately assessing the doctor's workload. It should also be noted that a higher workload index indicates a more saturated workload and less competitiveness, while a lower workload index indicates an unsaturated workload and the ability to compete for patients.

[0052] Step S432: Estimate the processing time for the number of patients to obtain the task processing time; predict the processing time for the medical items in the predicted medical node to obtain the time for new tasks; sum the task processing time and the time for new tasks; and calculate the ratio of the sum to the idle time to obtain the predicted time pressure value.

[0053] In some possible embodiments of this application, the task processing time corresponding to the number of patients is estimated based on historical medical records. For example, different doctors may require different, but roughly the same, times to diagnose a particular disease. The average time for diagnosing the disease by different doctors is taken, thus each patient's data corresponds to a diagnosis time. Summing these times yields the task processing time corresponding to the number of patients. Similarly, the processing time for the predicted medical items in the predicted medical node is predicted based on historical medical records, resulting in additional time. Then, the task processing time is summed with the additional task time to obtain the estimated time required to process all tasks. The ratio of the estimated time to the idle time yields the predicted time pressure value. This predicted time pressure value reflects the busyness of members at the predicted patient medical node, preparing for future increases or decreases in competitiveness. When the predicted time pressure value is greater than 1, it indicates that members are overworked at that time; when the predicted time pressure value is less than or equal to 1, it indicates that members can still compete for patients.

[0054] Step S433: Match the consultation time and idle time in the predicted consultation node to obtain the time matching degree.

[0055] In some possible embodiments of this application, the predicted appointment time in the appointment node is matched with the idle time. Specifically, a time window sliding method can be used for matching, sliding the appointment time in the predicted appointment node towards future time periods according to the size of the time window. When it completely overlaps with the idle time, the time matching degree is 1; when it partially overlaps with the idle time (including boundary overlap), the time matching degree is 0.5; when it does not overlap with the idle time at all, the time matching degree is 0. It should be noted that a higher time matching degree indicates that the workload of the members has not yet been overloaded.

[0056] Step S434: Calculate the penalty for the load index, time pressure forecast, and time matching degree to obtain the bidding penalty score.

[0057] In some possible embodiments of this application, a penalty calculation is performed on the load index, the predicted time pressure, and the time matching degree to obtain a bidding penalty score, including but not limited to: using a preset bidding penalty formula to calculate the penalty on the load index, the predicted time pressure, and the time matching degree to obtain a bidding penalty score; The formula for bidding penalty is expressed as follows: Penalty=δ*(1-TWM)+(1-δ)*max(1,α*RWI+β*TPF(t)) Where Penalty represents the bidding penalty score, TWM represents the time matching degree, RWI represents the load index, TPF(t) represents the time pressure forecast value, max() represents the maximization function, and δ, α, and β all represent adjustment parameters.

[0058] Specifically, during busy periods, the penalty is increased; the higher the bidding penalty score, the lower the sign of the penalty score after the above calculation. This facilitates subsequent summation calculations, reducing the bidding score and decreasing competition. The degree of reduction is estimated from three dimensions: load index, predicted time pressure, and time matching degree. In the above formula, the larger the load index, the larger the RWI, and the smaller the TWM value. The larger the 1-TWM value, the larger the TPF(t). The busier the member, the larger the value of α*RWI+β*TPF(t). Under heavy workload conditions, α*RWI+β*TPF(t) is greater than 1, so the maximum value is taken. Under light workload conditions, α*RWI+β*TPF(t) is less than 1, so the maximum value of 1 is taken through the maximization function to avoid a large number of competing patients and increased workload. The calculation shows that when a member is busy, the larger the bidding penalty score (negative sign), the smaller the value, and the lower the probability of winning the bid in subsequent bidding processes.

[0059] For example, in the case of non-compliance, δ is 0.6, α is 0.4, β is 0.6, TWM represents a time matching degree of 0, RWI represents a load index of 0.7, TPF(t) represents a predicted time stress value of 0.8, and Penalty = 0.6*(1-0) + 0.4*max(1, 0.4*0.7 + 0.6*0.8) = 1.6, which can avoid a large number of competing patients.

[0060] Step S440: Perform nonlinear game processing on the bidding reward score and bidding penalty score to obtain the game result.

[0061] In one embodiment, the bidding reward score and bidding penalty score are subjected to nonlinear game processing to obtain the game result, including but not limited to: The bidding reward score and bidding penalty score are subjected to nonlinear game processing using a preset game formula to obtain the game result; The game theory formula is expressed as follows: Where R represents the bidding reward score, P represents the bidding penalty score, and λ, φ, and c are adjustment parameters. represents the historical average fitness value, ni represents the number of times it was selected in history, t represents the total number of bids in history, and V represents the game outcome.

[0062] In some possible embodiments of this application, the higher the professional level of a member, the more advantageous their bidding. The bidding reward score is used as the numerator, and the bidding penalty score as the denominator. A larger numerator and a smaller denominator result in a higher game outcome and greater competitiveness. In the above formula, λ is 0.5, φ is 0.3, and a confidence level cap is added to dynamically adjust weight priorities. If the value of P is too high, the denominator exponentially increases explosively, forcibly lowering the priority. This process continuously explores and selects more suitable members. By applying non-linear game processing to the bidding reward score and bidding penalty score, the competitiveness of qualified members can be improved, achieving maximum fit and thus improving the efficiency of medical services. It should be noted that UCB... i This was obtained through statistical analysis of historical data, which will not be elaborated upon here. It should also be noted that the sign of the bidding penalty score is not included in the calculation during the game.

[0063] Step S450: Obtain the first weight corresponding to the bidding score, the second weight corresponding to the bidding reward score, and the third weight corresponding to the bidding penalty score, wherein the sum of the first weight, the second weight, and the third weight is 1.

[0064] In one embodiment, the first weight, second weight, and third weight are all determined by professionals integrating actual recorded data and combining it with experience. The settings are stored for subsequent adjustments and data retrieval. If the actual situation changes, the first weight, second weight, and third weight can be adjusted. For example, the first weight is 0.5, the second weight is 0.3, and the third weight is 0.2. The first weight corresponding to the bidding score, the second weight corresponding to the bidding bonus score, and the third weight corresponding to the bidding penalty score are obtained through a preset read function, so that the suitability bid can be calculated subsequently based on the first weight, second weight, and third weight. The preset read function can be either the `read()` function or the `open()` function.

[0065] Step S460: Adjust the second and third weights using the game results. Sum the results of multiplying the first weight by the bidding score, the adjusted second weight by the bidding reward score, and the adjusted third weight by the bidding penalty score to obtain the fit bid.

[0066] like Figure 4 As shown, the second and third weights are adjusted based on the game results, including but not limited to the following steps: Step S461: If the game result is greater than the preset adjustment threshold, increase the second weight, decrease the third weight, and keep the sum of the second weight and the third weight unchanged. Step S462: If the game result is less than the preset adjustment threshold, reduce the second weight, increase the third weight, and keep the sum of the second weight and the third weight unchanged. Step S463: If the game result is equal to the preset adjustment threshold, set the second weight and the third weight to the same value.

[0067] In some possible embodiments of this application, the preset adjustment threshold can be 0.5. If the game result is greater than the preset adjustment threshold, it indicates that the bidding reward score is dominant. Therefore, the second weight corresponding to the bidding reward score is increased, and the third weight corresponding to the bidding penalty score is decreased. Furthermore, the sum of the second and third weights remains unchanged, so that the sum of the first, second, and third weights is 1. For example, the second weight was originally 0.3, and the third weight was originally 0.2. After adjustment, the second weight is 0.4, and the third weight is 0.1.

[0068] If the game result is less than the preset adjustment threshold, the second weight is reduced and the third weight is increased. The sum of the second and third weights remains unchanged, indicating that the bidding penalty score is dominant and members are relatively busy. Therefore, the second weight corresponding to the bidding reward score is reduced and the third weight corresponding to the bidding penalty score is increased, while ensuring that the sum of the second and third weights remains unchanged, so that the sum of the first, second, and third weights is 1. If the game result is equal to the preset adjustment threshold, it indicates that the bidding reward score and the bidding penalty score are equally dominant. The second and third weights are set to the same value, and the sum of the first, second, and third weights is 1. The second and third weights are adjusted according to the game result to dynamically adjust the bidding score in the future, improving the competitiveness of eligible members. The sum of the product of the first weight and the bidding score, the product of the adjusted second weight and the bidding reward score, and the product of the adjusted third weight and the bidding penalty score is calculated to obtain the fitness bid, which is the optimal fitness result.

[0069] In one embodiment, the suitability bids are sorted, for example, in descending order. The member corresponding to the highest-ranked suitability bid is selected and pushed to the patient. Selecting members to push to patients in this way not only allows for patient tracking but also effectively coordinates medical resources, rationally allocating medical resources during peak demand periods and improving the efficiency of medical services.

[0070] like Figure 5As shown in the figure, this application embodiment provides a doctor resource allocation device 100. The doctor resource allocation device 100 acquires the patient's medical records through a data acquisition module 110. The medical records include the doctor's diagnostic score and the attending physician. The doctor's diagnostic score is obtained by the doctor scoring the patient's condition during the consultation. The content recorded in the medical records provides data support for subsequent calculations. A data prediction module 120 uses a preset target prediction model to predict the consultation node based on the medical records, obtaining a predicted consultation node. This prediction allows for the matching of appropriate members to the predicted consultation node, improving service efficiency. Then, a data grouping module 130 assigns the patient to the consultation group of the attending physician. Members of the physician's group have high professional and work relevance, providing support for subsequent resource allocation. The device acquires data from each member of the consultation group. Each member's professional level, work status, and bidding score are considered. Members in the consultation group have access to view patients' medical records, allowing other members in the group to understand the patient's condition. This alleviates the patient tracking pressure on attending physicians and avoids overburdening them. The bidding score is calculated by members in the consultation group using a preset bidding function based on their professional level, doctor's diagnostic score, work status, and predicted consultation time. Members submit their bidding scores to compete for patients, avoiding waste of doctor resources. The data matching module 140 dynamically adjusts the bidding score using professional level, predicted consultation time, doctor's diagnostic score, and work status to obtain a suitable bid. Members with the highest suitable bid are selected and pushed to the patient. Through dynamic adjustment, more suitable members can be matched to patients, improving the efficiency of medical services.

[0071] It should be noted that the data acquisition module 110 is connected to the data prediction module 120, the data prediction module 120 is connected to the data grouping module 130, and the data grouping module 130 is connected to the data matching module 140. The above-mentioned doctor resource allocation method is applied to the doctor resource allocation device 100. The doctor resource allocation device 100 acquires the patient's medical records, which include the doctor's diagnostic score and the attending physician. The doctor's diagnostic score is obtained by the doctor based on the severity of the patient's condition during the consultation. The content recorded in the medical records provides data support for subsequent calculations. Based on the medical records, a preset target prediction model is used to predict the consultation node, resulting in a predicted consultation node. This prediction allows for the matching of appropriate members to the predicted consultation node, improving service efficiency. Patients are assigned to the consultation group of the attending physician. Members of the doctor's group have high professional and job relevance, providing support for subsequent resource allocation. The device acquires the professional level and job information of each member in the consultation group. The system includes a status and bidding score. Members in the patient visit group have access to view patients' medical records, allowing other members in the group to understand the patient's condition. This alleviates the patient tracking pressure on attending physicians and avoids overburdening them. The bidding score is calculated by members in the patient visit group using a preset bidding function based on their professional level, doctor's diagnostic score, work status, and predicted visit time. Members bid for patients, avoiding waste of doctor resources. The bidding score is dynamically adjusted based on professional level, predicted visit time, doctor's diagnostic score, and work status to obtain a suitable bid. Members with the highest suitable bid are selected and sent to patients. Through dynamic adjustment, more suitable members can be matched to patients, improving the efficiency of medical services.

[0072] It should also be noted that the apparatus provided in the above embodiments is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and their specific implementation process can be found in the method embodiments, which will not be repeated here.

[0073] This application also discloses an electronic device. (See reference...) Figure 6 , Figure 6 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. The electronic device 500 may include: at least one processor 501, at least one network interface 504, a user interface 503, a memory 505, and at least one communication bus 502.

[0074] The communication bus 502 is used to enable communication between these components.

[0075] The user interface 503 may include a display screen and a camera. Optionally, the user interface 503 may also include a standard wired interface and a wireless interface.

[0076] The network interface 504 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0077] The processor 501 may include one or more processing cores. The processor 501 connects to various parts of the server using various interfaces and lines, and performs various server functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in memory 505, and by calling data stored in memory 505. Optionally, the processor 501 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 501 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content required for display; and the modem handles wireless communication. It is understood that the modem may also not be integrated into the processor 501 and may be implemented as a separate chip.

[0078] The memory 505 may include random access memory (RAM) or read-only memory. Optionally, the memory 505 may include a non-transitory computer-readable storage medium. The memory 505 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 505 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory 505 may also be at least one storage device located remotely from the aforementioned processor 501. (Refer to...) Figure 6 The memory 505, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and an application program for a doctor resource allocation method.

[0079] exist Figure 6 In the illustrated electronic device 500, the user interface 503 is mainly used to provide an input interface for the user and to acquire user input data; while the processor 501 can be used to call an application program stored in the memory 505 for a doctor resource allocation method. When executed by one or more processors 501, the electronic device 500 performs one or more methods as described in the above embodiments. It should be noted that, for the foregoing method embodiments, for the sake of simplicity, they are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, because according to this application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0080] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0081] In the various embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some service interface; the indirect coupling or communication connection between apparatuses or units may be electrical or other forms.

[0082] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0083] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0084] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0085] The above are merely exemplary embodiments of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. Other embodiments of this disclosure will readily conceive of those skilled in the art upon consideration of the specification and the disclosure of practical truths.

[0086] This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described in this disclosure. The specification and embodiments are to be considered exemplary only, and the scope and spirit of this disclosure are defined by the claims.

Claims

1. A method for allocating doctor resources, characterized in that, The method includes: Obtain the patient's medical records, which include the doctor's diagnostic score and the attending physician. The doctor's diagnostic score is obtained by the doctor during the consultation based on the severity of the patient's condition. Based on the medical records, the medical visit node is predicted using a preset target prediction model to obtain the predicted medical visit node. The patient is assigned to the consultation group of the attending physician, and the professional level, work status and bidding score of each member in the consultation group are obtained. The members in the consultation group have the right to view the patient's medical records. The bidding score is calculated by the members in the consultation group using a preset bidding function based on the professional level, the doctor's diagnosis score, the work status and the predicted consultation node. The bidding score is dynamically adjusted using the professional level, the predicted visit node, the doctor's diagnosis score, and the work status to obtain the suitability bid. Members matching the maximum suitability bid are then selected and pushed to the patient.

2. The method according to claim 1, characterized in that, The step of dynamically adjusting the bidding score using the professional level, the predicted consultation node, the doctor's diagnostic score, and the work status to obtain a suitable bid includes: when the doctor's diagnostic score is greater than a preset score threshold, quantifying the professional level and the consulting doctor's score respectively to obtain the level quantified score corresponding to the professional level and the consultation quantified score corresponding to the consulting doctor; summing the level quantified score and the consultation quantified score, and using the sum as the bidding bonus score; The bidding penalty score is calculated based on the working status and the predicted medical visit node, and the sign of the bidding penalty score is set to a negative number. The bidding reward score and the bidding penalty score are subjected to a nonlinear game to obtain the game result. Obtain the first weight corresponding to the bidding score, the second weight corresponding to the bidding reward score, and the third weight corresponding to the bidding penalty score, wherein the sum of the first weight, the second weight, and the third weight is 1; The second weight and the third weight are adjusted using the game results. The results of multiplying the first weight by the bidding score, the adjusted second weight by the bidding reward score, and the adjusted third weight by the bidding penalty score are summed to obtain the fit bid.

3. The method according to claim 2, characterized in that, The nonlinear game processing of the bidding reward score and the bidding penalty score to obtain the game result includes: The bidding reward score and the bidding penalty score are subjected to nonlinear game processing using a preset game formula to obtain the game result; The game theory formula is expressed as follows: Where R represents the bidding reward score, P represents the bidding penalty score, and λ, μ, and c are adjustment parameters. represents the historical average fitness value, ni represents the number of times it was selected in history, t represents the total number of bids in history, and V represents the game outcome.

4. The method according to claim 2, characterized in that, The work status includes idle time, number of patients received, and the proportion of urgent tasks. The calculation of the bidding penalty score based on the working status and the predicted visit node includes: The ratio of the number of patients received to a preset threshold for the number of patients received is calculated to obtain the saturation of the number of patients received. The saturation of the number of patients received is then weighted and summed with the proportion of emergency tasks to obtain the load index. The processing time for the number of patients is estimated to obtain the task processing time. The processing time for the medical items in the predicted medical node is predicted to obtain the new task time. The task processing time and the new task time are summed. The sum is compared with the idle time to obtain the predicted time pressure value. The time matching degree is obtained by matching the appointment time in the predicted appointment node with the idle time; The load index, the predicted time pressure, and the time matching degree are penalized to obtain the bidding penalty score.

5. The method according to claim 4, characterized in that, The penalty calculation for the load index, the predicted time pressure value, and the time matching degree to obtain the bidding penalty score includes: The load index, the predicted time pressure, and the time matching degree are penalized using a preset bidding penalty formula to obtain a bidding penalty score. The bidding penalty formula is expressed as follows: Penalty=δ*TWM+(1-δ)*max(1,α*RWI+β*TPF(t)) Wherein, Penalty represents the bidding penalty score, TWM represents the time matching degree, RWI represents the load index, TPF(t) represents the predicted time pressure value, max() represents the maximization function, and δ, α, and β all represent adjustment parameters.

6. The method according to claim 2, characterized in that, The adjustment of the second weight and the third weight using the game result includes: If the game result is greater than a preset adjustment threshold, the second weight is increased and the third weight is decreased, while the sum of the second weight and the third weight remains unchanged. If the game result is less than a preset adjustment threshold, the second weight is reduced, the third weight is increased, and the sum of the second weight and the third weight remains unchanged. If the game result is equal to the preset adjustment threshold, the second weight and the third weight are set to the same value.

7. The method according to claim 1, characterized in that, The target prediction model is obtained in the following way: Obtain multiple historical medical records, each of which includes multiple historical medical items, the historical medical time corresponding to each historical medical item, the next medical item corresponding to the historical medical item, and the next medical time corresponding to the next medical item. Multiple medical visits and multiple historical medical visits are input into a preset initial prediction model in a preset batch to obtain predicted medical visits and predicted medical visits. Calculate the values ​​of the loss functions for the predicted medical visit items, the predicted medical visit time, the next node medical visit items, and the next node medical visit time. Use the values ​​of the loss functions to train the initial prediction model in reverse until the preset training conditions are met, and obtain the target prediction model.

8. A doctor resource allocation device, characterized in that, The device includes: The data acquisition module (110) is used to acquire the patient's medical records, which include the doctor's diagnostic score and the attending physician. The doctor's diagnostic score is obtained by the doctor scoring the patient's condition during the consultation. The data prediction module (120) is used to predict the medical visit node based on the medical visit record using a preset prediction model, and obtain the predicted medical visit node. The data grouping module (130) is used to assign the patient to the consultation group of the attending physician, and to obtain the professional level, work status and bidding score of each member in the consultation group. The members in the consultation group have the right to view the patient's consultation record. The bidding score is calculated by the members in the consultation group using a preset bidding function based on the professional level, the doctor's diagnosis score, the work status and the predicted consultation node. The data matching module (140) is used to dynamically adjust the bidding score using the professional level, the predicted visit node, the doctor's diagnosis score, and the work status to obtain the suitability bid, and select members that match the maximum suitability bid to push to the patient.

9. An electronic device, characterized in that, The device includes a processor (501), a memory (505), a user interface (503), a communication bus (502), and a network interface (504). The processor (501), the memory (505), the user interface (503), and the network interface (504) are respectively connected to the communication bus (502). The memory (505) is used to store instructions. The user interface (503) and the network interface (504) are used to communicate with other devices. The processor (501) is used to execute the instructions stored in the memory (505) so that the electronic device (500) performs the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that, when executed, perform the method as described in any one of claims 1-7.

Images