Guided triage management method and device based on smart medical care

Through the smart medical guidance and triage management method, the pre-training model is used to match and evaluate doctors, the waiting problem caused by the large number of doctors registered in hospitals in big cities is solved, and the patient's medical experience and resource utilization efficiency are improved.

CN118471448BActive Publication Date: 2024-11-15GUANGDONG QUNCHUANG INFORMATION TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410550484.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-05-06
Publication Date
2024-11-15
Estimated Expiration
2044-05-06

AI Technical Summary

Technical Problem

In Grade A hospitals in big cities, doctors have a large number of registered users during certain periods, which leads to a long queue for patients to see the doctor after picking up their numbers, resulting in a low patient experience.

Method used

Using a guided triage management method based on smart medical care, the patient's symptom information and medical treatment period are received through cloud servers, and the doctors that match the disease field are matched, and the pre-trained patient number prediction model and medical experience scoring model are used to predict the number of patients and the number of patients to be treated and the number of patients is predicted. Evaluate medical experience and optimize medical treatment process to reduce waiting time.

Benefits of technology

It improves the patient's medical experience, shortens the queue time, makes rational use of medical resources, and improves the patient's satisfaction with the treatment process.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118471448B_ABST
    Figure CN118471448B_ABST
Patent Text Reader

Abstract

The present application provides a guidance and triage management method and device based on smart medical care, which relates to the field of smart treatment technology. The first patient number change characteristics of each doctor in the target medical treatment period and the time length from the first moment to the target medical treatment period are input into the first patient number prediction model to obtain the first predicted patient number, and the second patient number change characteristics of each doctor in multiple consecutive historical medical treatment periods with the same duration as the target medical treatment period in the second time period are input into the second patient number prediction model to obtain the second predicted patient number; the first predicted patient number and the second patient number corresponding to each doctor are weighted summed to obtain the final predicted patient number of each doctor in the target medical treatment period; based on the location information of the user terminal, the location information of the tertiary hospitals to which the selected doctors ranked in the top N in medical experience scores belong, and the medical treatment process of the selected doctors, medical guidance and triage information is generated.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of intelligent treatment technology, and in particular to a guided diagnosis and triage management method and device based on intelligent medical care. Background Art

[0002] With the development of technology, patients can remotely select a target doctor on the user terminal to register for a target time period (such as 9:00-10:00; or 10:00-11:00, etc.). Then, patients can arrive at the hospital to get a number during the target time period, and queue up to see a doctor in the order of the number they got. For some tertiary hospitals in large cities, some doctors (especially those who treat seasonal diseases) have a large number of registrations during certain time periods, resulting in a long waiting time for patients to wait in line for medical treatment after getting a number, which is a waste of time and a low patient experience. Summary of the invention

[0003] The present application provides a guidance and triage management method and device based on smart medical care, which is used to solve the problem of low patient experience in seeing a doctor in the prior art.

[0004] In the first aspect, the present application provides a guidance and triage management method based on smart medical care, which is applied to a cloud server. The method provided by the present application includes:

[0005] At a first moment, receiving symptom information transmitted by a patient based on a user terminal and a target medical consultation period on a target date;

[0006] Acquire the disease domain of the patient's disease based on the symptom information, and when it is determined that the disease domain is a seasonal disease, determine a doctor matching the disease domain in a tertiary hospital in the target area;

[0007] For each doctor matching the disease field, obtain the number of first historical patients in the target medical treatment period in multiple historical days in the first time period before the first moment, and obtain the number of second historical patients in multiple consecutive historical medical treatment periods with the same duration as the target medical treatment period in the second time period before the first moment;

[0008] According to the first historical number of patients visiting the target medical treatment period in each of the multiple historical days, the first patient number change characteristics of each doctor matching the disease field in the target medical treatment period are extracted, and according to the second historical number of patients visiting the target medical treatment period in multiple consecutive historical medical treatment periods with the same duration as the target medical treatment period, the second patient number change characteristics of each doctor matching the disease field in multiple consecutive historical medical treatment periods with the same duration as the target medical treatment period in the second time period are extracted;

[0009] Input the first patient quantity change characteristics of each doctor matching the disease field in the target medical treatment period and the time length from the first moment to the target medical treatment period into the pre-trained first patient quantity prediction model to obtain the first predicted patient quantity output by the first patient quantity prediction model, wherein the first patient quantity prediction model is obtained by inputting a training sample set consisting of the first historical patient quantity change characteristics of multiple historical target medical treatment periods of each doctor, the historical time length from the historical first moment to the corresponding historical target medical treatment period, and the first actual number of patients visiting the corresponding historical target medical treatment period into the first neural network for training, and the historical first moment is the moment of receiving the historical symptom information and the historical medical treatment period on the historical target date;

[0010] Input the second patient quantity change characteristics of multiple consecutive historical medical treatment periods of each doctor matched with the disease field within the second time period and equal to the target medical treatment period into the pre-trained second patient quantity prediction model to obtain the second predicted patient quantity output by the second patient quantity prediction model, where the second time period is before the first moment and continuous with the first moment; the second patient quantity prediction model is obtained by inputting the second historical patient quantity change characteristics of multiple consecutive historical medical treatment periods of each doctor within the historical second time period and equal to the target medical treatment period and the corresponding second actual number of patients as a training sample set into the second neural network for training, wherein the historical second time period is before the historical first moment and continuous with the historical first moment, and the historical first moment is the moment when the historical symptom information and the historical medical treatment period on the historical target date are received;

[0011] The weighted sum of the first predicted number of patients and the second number of patients corresponding to each doctor is performed to obtain the final predicted number of patients for each doctor in the target medical treatment period;

[0012] Get the expertise of each doctor matching the condition area;

[0013] Obtain the medical treatment distance between the user terminal's location information and the preset location information of the top three hospitals to which each doctor belongs;

[0014] The final predicted number of patients for each doctor in the target medical treatment period, the professional level of each doctor matching the disease field, and the medical treatment distance between the user terminal's location information and the preset location information of the Class-A tertiary hospital to which each doctor belongs are input into the pre-trained medical treatment experience scoring model, and the medical treatment experience score for each doctor is output, wherein the medical treatment experience score is negatively correlated with the final predicted number of patients, positively correlated with the professional level of the doctor matching the disease field, and negatively correlated with the medical treatment distance; wherein the medical treatment experience scoring model is obtained by inputting multiple historical final predicted numbers of patients, corresponding historical doctor's professional levels, and corresponding historical medical treatment distances as a training sample set into the third neural network for training;

[0015] Select the top N doctors in terms of medical experience scores and transmit them to the user terminal for display;

[0016] Receive a selected doctor from the top N doctors selected by the user sent by the user terminal;

[0017] Based on the location information of the user terminal, the location information of the tertiary hospital to which the selected doctor belongs, and the preset medical process of the selected doctor, medical guidance and triage information is generated and sent to the user terminal for display.

[0018] In the second aspect, the present application provides a guidance and triage management device based on smart medical care, which is configured on a cloud server. The device provided by the present application includes:

[0019] A data receiving unit, configured to receive, at a first moment, symptom information of a patient transmitted based on a user terminal and a target medical treatment period on a target date;

[0020] A data acquisition unit, for acquiring the symptom field of the disease suffered by the patient according to the symptom information, and when it is determined that the symptom field is a seasonal symptom, determining a doctor matching the symptom field in a tertiary hospital in a target area;

[0021] The data acquisition unit is further configured to acquire, for each doctor matching the disease field, a first historical number of patients visiting the target medical treatment period in a plurality of historical days in a first time period before the first moment, and acquire a second historical number of patients visiting the target medical treatment period in a plurality of consecutive historical medical treatment periods in a second time period before the first moment and having a duration equal to that of the target medical treatment period;

[0022] A feature extraction unit is used to extract the first patient quantity change characteristics of each doctor matching the disease field in the target medical treatment period according to the first historical patient quantity of each target medical treatment period in a plurality of historical days, and to extract the second patient quantity change characteristics of each doctor matching the disease field in a plurality of consecutive historical medical treatment periods with a duration equal to that of the target medical treatment period within the second time period according to the second historical patient quantity of each doctor matching the disease field;

[0023] a data prediction unit, for inputting the first patient quantity change characteristics of each doctor matching the disease field in the target medical treatment period and the duration from the first moment to the target medical treatment period into a pre-trained first patient quantity prediction model, so as to obtain a first predicted patient quantity output by the first patient quantity prediction model, wherein the first patient quantity prediction model is obtained by inputting a training sample set consisting of the first historical patient quantity change characteristics of multiple historical target medical treatment periods of each doctor, the historical duration from the historical first moment to the corresponding historical target medical treatment period, and the first actual number of patients visiting the corresponding historical target medical treatment period into a first neural network for training, wherein the historical first moment is the moment of receiving historical symptom information and the historical medical treatment period on the historical target date;

[0024] The data prediction unit is used to input the second patient quantity change characteristics of multiple consecutive historical medical treatment periods of each doctor matching the disease field within a second time period and equal to the duration of the target medical treatment period into the pre-trained second patient quantity prediction model to obtain the second predicted patient quantity output by the second patient quantity prediction model, wherein the second time period is before the first moment and continuous with the first moment; the second patient quantity prediction model is obtained by inputting the second historical patient quantity change characteristics of multiple consecutive historical medical treatment periods of each doctor within the historical second time period and equal to the duration of the target medical treatment period and the corresponding second actual number of patients as a training sample set into the second neural network for training, wherein the historical second time period is before the historical first moment and continuous with the historical first moment, and the historical first moment is the moment when the historical symptom information and the historical medical treatment period on the historical target date are received;

[0025] a data weighting unit, configured to perform a weighted summation of the first predicted number of patients and the second number of patients corresponding to each doctor, so as to obtain a final predicted number of patients for each doctor in the target medical treatment period;

[0026] The data acquisition unit is further used to acquire the professional level of each doctor matching the disease field;

[0027] The data acquisition unit is further used to acquire the medical treatment distance between the location information of the user terminal and the location information of the preset tertiary hospitals to which each doctor belongs;

[0028] A data output unit is used to input the final predicted number of patients of each doctor in the target medical treatment period, the professional level of each doctor matching the disease field, and the medical treatment distance between the location information of the user terminal and the location information of the preset tertiary hospital to which each doctor belongs, into a pre-trained medical treatment experience scoring model, and output the medical treatment experience score of each doctor, wherein the medical treatment experience score is negatively correlated with the final predicted number of patients, positively correlated with the professional level of the doctor matching the disease field, and negatively correlated with the medical treatment distance; wherein the medical treatment experience scoring model is obtained by inputting a plurality of historical final predicted numbers of patients, corresponding historical professional levels of doctors, and corresponding historical medical treatment distances as a training sample set into a third neural network for training;

[0029] A data transmission unit, used to select the top N doctors in the medical experience ranking and transmit them to the user terminal for display;

[0030] The data receiving unit is further used to receive a selected doctor from the top N doctors selected by the user from the user terminal;

[0031] A data sending unit is used to generate medical guidance and triage information based on the location information of the user terminal, the location information of the tertiary hospital to which the selected doctor belongs, and the preset medical treatment process of the selected doctor, and send the medical guidance and triage information to the user terminal for display.

[0032] Compared with the prior art, this application has the following technical effects:

[0033] The present application provides a guided triage management method and device based on smart medical care, which can input the first patient quantity change characteristics of each doctor matched with the disease field in the target medical treatment period and the time length from the first moment to the target medical treatment period into the pre-trained first patient quantity prediction model to obtain the first predicted patient quantity output by the first patient quantity prediction model. Since the first patient quantity prediction model is obtained by inputting the first historical patient quantity change characteristics of multiple historical target medical treatment periods of each doctor, the historical time length from the historical first moment to the corresponding historical target medical treatment period, and the first actual number of patients in the corresponding historical target medical treatment period into the first neural network for training, the obtained first actual number of patients has high accuracy;

[0034] The second patient quantity change characteristics of multiple consecutive historical medical treatment periods of each doctor matching the disease field and equal to the target medical treatment period in the second time period can also be input into the pre-trained second patient quantity prediction model to obtain the second predicted patient quantity output by the second patient quantity prediction model, where the second time period is before the first moment and continuous with the first moment; since the second patient quantity prediction model is obtained by inputting the second historical patient quantity change characteristics of multiple consecutive historical medical treatment periods of each doctor in the historical second time period and equal to the target medical treatment period and the corresponding second actual number of patients as the training sample set into the second neural network for training, the second actual number of patients obtained is highly accurate;

[0035] Since the accuracy of the first actual number of patients visiting the hospital is high and the accuracy of the second actual number of patients visiting the hospital is high, a weighted sum is performed on the first predicted number of patients and the second number of patients corresponding to each doctor, and the final predicted number of patients for each doctor in the target medical treatment period has a high accuracy.

[0036] In addition, the professional level of each doctor matching the disease field is obtained, and the medical treatment distance between the location information of the user terminal and the location information of the tertiary hospitals to which each doctor belongs is obtained;

[0037] The final predicted number of patients of each doctor in the target medical treatment period, the professional level of each doctor matching the disease field, and the medical treatment distance between the user terminal's location information and the preset location information of the Class-A tertiary hospital to which each doctor belongs are input into the pre-trained medical treatment experience scoring model, and the medical treatment experience score for each doctor is output. It can be understood that since the medical treatment experience score output by the medical treatment experience scoring model takes into account the three dimensions of the final predicted number of patients of each doctor in the target medical treatment period, the professional level of each doctor matching the disease field, and the medical treatment distance between the user terminal's location information and the preset location information of the Class-A tertiary hospital to which each doctor belongs, with high accuracy, the obtained medical treatment experience score has high reliability;

[0038] Finally, based on the location information of the user terminal, the location information of the tertiary hospitals to which the top N selected doctors in the medical experience score ranking belong, and the preset medical process of the selected doctors, medical guidance and triage information is generated and sent to the user terminal for display. In this way, the patient's waiting time for medical treatment is shortened, the medical distance is shortened, and the medical treatment effect is good, which greatly improves the patient's experience of the entire treatment process, effectively and accurately diverts patients, and makes more reasonable use of medical resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0040] Figure 1 A flowchart of a smart medical guidance and triage management method provided in an embodiment of the present application;

[0041] Figure 2 A functional module block diagram of the guidance and triage management device based on smart medical care provided in an embodiment of the present application. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments made by ordinary technicians in this field under the enlightenment of the embodiments belong to the scope of protection of the present application.

[0043] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0044] See also Figure 1 The embodiment of the present application provides a guidance and triage management method based on smart medical care, which is applied to a cloud server and includes:

[0045] S101: receiving, at a first moment, symptom information of a patient transmitted from a user terminal and a target medical consultation period on a target date.

[0046] Among them, the target medical treatment period of the target date can be 10:00-11:00 on May 1, 2024; it can also be 14:00-15:00 on May 10, 2024, etc.

[0047] S102: When the symptom field of the disease suffered by the patient is obtained based on the symptom information and when it is determined that the symptom field is a seasonal disease, a doctor matching the symptom field in a tertiary hospital in the target area is determined.

[0048] Exemplarily, the symptom information can be input into a pre-trained disease domain recognition model to determine the disease domain of the patient's disease; wherein the disease domain recognition model is trained by inputting historical symptom information and the corresponding actual disease domain into the network to be trained, and can determine whether the disease domain is a seasonal disease by searching. Seasonal diseases include but are not limited to pneumonia, colds, chronic bronchitis, cor pulmonale, asthma, myocardial infarction, mental illness, etc.

[0049] S103: For each doctor matching the disease field, obtain the number of first historical patients visiting the target medical treatment period in multiple historical days within the first time period before the first moment, and obtain the number of second historical patients visiting the target medical treatment period in multiple consecutive historical medical treatment periods within the second time period before the first moment and with the same duration as the target medical treatment period.

[0050] For example, when the first moment is 9:00 on April 30, 2024, the target medical treatment period of the target date is 10:00-11:00 on May 1, 2024, and the first time period is 1 month, the first historical number of patients visiting the target medical treatment period 10:00-11:00 between 9:00 on March 31, 2024 and 9:00 on April 30, 2024 is obtained.

[0051] When the first moment is 9:00 on April 30, 2024, the target medical treatment period on the target date is 10:00-11:00 on May 1, 2024, and the second time period is half a month, multiple consecutive historical medical treatment periods between 9:00 on April 15, 2024 and 9:00 on April 30, 2024 are obtained, and the duration of the target medical treatment period is equal to that of the target medical treatment period (such as 9:00-10:00; 10:00-11:00; 11:00-12:00, etc. every day). It should be noted that the historical medical treatment period is a sub-time period within the working time period (9:00-6:00).

[0052] S104: Extract the first patient quantity change characteristics of each doctor matching the disease field in the target medical treatment period based on the first historical patient quantity of each target medical treatment period in multiple historical days, and extract the second patient quantity change characteristics of each doctor matching the disease field in multiple consecutive historical medical treatment periods with the same duration as the target medical treatment period in the second time period based on the second historical patient quantity of each target medical treatment period in each historical medical treatment period.

[0053] Specifically, S104 can be implemented as follows: according to the first deep confidence DBN network, the first time-series dependency of the first historical number of patients visiting each target medical period in multiple historical days is extracted, and the first historical number of patients visiting each target medical period in multiple historical days and the first dependency are integrated to obtain the first patient number change characteristics of each doctor matching the disease field in the target medical period; according to the second deep confidence DBN network, the second time-series dependency of the second historical number of patients visiting each historical medical period with a duration equal to the target medical period is extracted, and the second historical number of patients visiting each historical medical period with a duration equal to the target medical period is integrated to obtain the second patient number change characteristics of each doctor matching the disease field in the second time period with a duration equal to the target medical period, which has high reliability.

[0054] S105: Input the first patient quantity change characteristics of each doctor matching the disease field during the target medical treatment period and the length of time from the first moment to the target medical treatment period into the pre-trained first patient quantity prediction model to obtain the first predicted patient quantity output by the first patient quantity prediction model.

[0055] Among them, the first patient quantity prediction model is obtained by inputting a training sample set consisting of the first historical patient quantity change characteristics of each doctor's multiple historical target medical time periods, the historical duration of the historical first moment from the corresponding historical target medical time period, and the first actual number of patients visiting the corresponding historical target medical time period into the first neural network for training. The historical first moment is the moment when the historical symptom information and the historical medical time period on the historical target date are received.

[0056] S106: Input the second patient quantity change characteristics of multiple consecutive historical medical treatment periods within the second time period and equal to the duration of the target medical treatment period for each doctor matching the disease field into the pre-trained second patient quantity prediction model to obtain the second predicted patient quantity output by the second patient quantity prediction model.

[0057] The second time period is before the first moment and continuous with the first moment; the second patient quantity prediction model is obtained by inputting the second historical patient quantity change characteristics of multiple consecutive historical medical treatment periods of each doctor within the historical second time period and equal to the duration of the target medical treatment period and the corresponding second actual number of patients as training sample sets into the second neural network for training.

[0058] The second historical time period is before the first historical moment and is continuous with the first historical moment, and the first historical moment is the moment when the historical symptom information and the historical medical treatment period on the historical target date are received.

[0059] S107: Perform a weighted summation of the first predicted number of patients and the second number of patients corresponding to each doctor to obtain the final predicted number of patients for each doctor in the target medical treatment period.

[0060] S108: Obtain the professional level of each doctor matching the disease field.

[0061] Specifically, S108 can be implemented as obtaining the working years, professional title, academic qualifications, and job title of each doctor matching the disease field; finding the first professional score corresponding to the working years, the second professional score corresponding to the professional title, the third professional score corresponding to the academic qualifications, and the fourth professional score corresponding to the job title; taking a weighted average of the first professional score, the second professional score, the third professional score, and the fourth professional score to obtain a comprehensive professional score, and using the comprehensive professional score as the professional level of the doctor matching the disease field. In this way, the professional level of the doctor matching the disease field can be made highly reliable.

[0062] S109: Obtaining the medical treatment distance between the location information of the user terminal and the location information of the preset tertiary hospitals to which each doctor belongs.

[0063] S110: Input the final predicted number of patients for each doctor during the target medical treatment period, the professional level of each doctor matching the disease field, and the medical treatment distance between the user terminal's location information and the preset location information of the tertiary hospitals to which each doctor belongs into the pre-trained medical treatment experience scoring model, and output the medical treatment experience score for each doctor.

[0064] Among them, the medical experience score is negatively correlated with the final predicted number of patients, positively correlated with the professional level of doctors matching the disease field, and negatively correlated with the medical distance. Among them, the medical experience scoring model is trained by inputting multiple historical final predicted numbers of patients, the corresponding historical doctor's professional level, and the corresponding historical medical distance as a training sample set into the third neural network.

[0065] S111: Select the top N doctors in terms of medical experience scores and transmit them to the user terminal for display.

[0066] S112: Receive a selected doctor from the top N ranked doctors selected by the user sent by the user terminal.

[0067] S113: Generate medical guidance and triage information based on the location information of the user terminal, the location information of the tertiary hospital to which the selected doctor belongs, and the preset medical treatment process of the selected doctor, and send the medical guidance and triage information to the user terminal for display.

[0068] Among them, the medical guidance and triage information may include but is not limited to the navigation path from the location information of the user terminal to the location information of the tertiary hospital to which the selected doctor belongs, the profile of the selected doctor, the target medical time period, the number of patients during the target medical time period, the medical process of the selected doctor (such as first making an examination report, then taking the pulse, then a simple physical examination, then asking about the condition, and finally giving instructions for medication, etc.), etc., and are not limited here.

[0069] In summary, the embodiment of the present application provides a guided triage management method based on smart medical care, which can input the first patient quantity change characteristics of each doctor matched with the disease field in the target medical treatment period and the time length from the first moment to the target medical treatment period into the pre-trained first patient quantity prediction model to obtain the first predicted patient quantity output by the first patient quantity prediction model. Since the first patient quantity prediction model is obtained by inputting the first historical patient quantity change characteristics of multiple historical target medical treatment periods of each doctor, the historical time length from the historical first moment to the corresponding historical target medical treatment period, and the first actual number of patients in the corresponding historical target medical treatment period into the first neural network for training, the obtained first actual number of patients has high accuracy;

[0070] The second patient quantity change characteristics of multiple consecutive historical medical treatment periods of each doctor matching the disease field and equal to the target medical treatment period in the second time period can also be input into the pre-trained second patient quantity prediction model to obtain the second predicted patient quantity output by the second patient quantity prediction model, where the second time period is before the first moment and continuous with the first moment; since the second patient quantity prediction model is obtained by inputting the second historical patient quantity change characteristics of multiple consecutive historical medical treatment periods of each doctor in the historical second time period and equal to the target medical treatment period and the corresponding second actual number of patients as the training sample set into the second neural network for training, the second actual number of patients obtained is highly accurate;

[0071] Since the accuracy of the first actual number of patients visiting the hospital is high and the accuracy of the second actual number of patients visiting the hospital is high, a weighted sum is performed on the first predicted number of patients and the second number of patients corresponding to each doctor, and the final predicted number of patients for each doctor in the target medical treatment period has a high accuracy.

[0072] In addition, the professional level of each doctor matching the disease field is obtained, and the medical treatment distance between the location information of the user terminal and the location information of the tertiary hospitals to which each doctor belongs is obtained;

[0073] The final predicted number of patients of each doctor in the target medical treatment period, the professional level of each doctor matching the disease field, and the medical treatment distance between the user terminal's location information and the preset location information of the Class-A tertiary hospital to which each doctor belongs are input into the pre-trained medical treatment experience scoring model, and the medical treatment experience score for each doctor is output. It can be understood that since the medical treatment experience score output by the medical treatment experience scoring model takes into account the three dimensions of the final predicted number of patients of each doctor in the target medical treatment period, the professional level of each doctor matching the disease field, and the medical treatment distance between the user terminal's location information and the preset location information of the Class-A tertiary hospital to which each doctor belongs, with high accuracy, the obtained medical treatment experience score has high reliability;

[0074] Finally, based on the location information of the user terminal, the location information of the tertiary hospitals to which the top N selected doctors in the medical experience score ranking belong, and the preset medical process of the selected doctors, medical guidance and triage information is generated and sent to the user terminal for display. In this way, the patient's waiting time for medical treatment is shortened, the medical distance is shortened, and the medical treatment effect is good, which greatly improves the patient's experience of the entire treatment process, effectively and accurately diverts patients, and makes more reasonable use of medical resources.

[0075] Furthermore, after S113, the method provided in the embodiment of the present application also includes: receiving a medical experience score sent by the patient based on the user terminal after the patient selects a doctor for medical treatment; determining the score difference between the received medical experience score and the medical experience score output by the medical experience scoring model; when the score difference is lower than the set difference threshold, updating the medical experience scoring model according to the score difference, so that the reliability of the medical experience scoring model can be made higher.

[0076] See also Figure 2 , the embodiment of the present application also provides a guidance and triage management device based on smart medical care, which is configured on a cloud server. It should be noted that the basic principles and technical effects of the guidance and triage management device based on smart medical care provided in the embodiment of the present application are the same as those in the above-mentioned embodiment. For the sake of brief description, for parts not mentioned in the embodiment of the present application, please refer to the corresponding contents in the above-mentioned embodiment. The device provided in the embodiment of the present application includes a data receiving unit, a data acquisition unit, a feature extraction unit, a data prediction unit, a data weighting unit, a data output unit, a data transmission unit, and a data sending unit, wherein,

[0077] The data receiving unit is used to receive the symptom information of the patient transmitted based on the user terminal and the target medical treatment period on the target date at a first moment.

[0078] The data acquisition unit is used to obtain the symptom field of the patient's disease based on the symptom information, and when it is determined that the symptom field is a seasonal disease, determine the doctor who matches the symptom field in the tertiary hospitals in the target area.

[0079] The data acquisition unit is also used to obtain, for each doctor matched with the disease field, the number of first historical patients visiting the target medical treatment period in multiple historical days within a first time period before the first moment, and to obtain the number of second historical patients visiting the target medical treatment period in multiple consecutive historical medical treatment periods with the same duration as the target medical treatment period within a second time period before the first moment.

[0080] The feature extraction unit is used to extract the first patient quantity change characteristics of each doctor matching the disease field in the target medical treatment period according to the first historical patient quantity of each target medical treatment period in multiple historical days, and to extract the second patient quantity change characteristics of each doctor matching the disease field in multiple consecutive historical medical treatment periods with the same duration as the target medical treatment period in a second time period according to the second historical patient quantity of multiple consecutive historical medical treatment periods with the same duration as the target medical treatment period.

[0081] The data prediction unit is used to input the first patient quantity change characteristics of each doctor matching the disease field in the target medical treatment period and the length of time from the first moment to the target medical treatment period into the pre-trained first patient quantity prediction model to obtain the first predicted patient quantity output by the first patient quantity prediction model.

[0082] Among them, the first patient quantity prediction model is obtained by inputting a training sample set consisting of the first historical patient quantity change characteristics of each doctor's multiple historical target medical time periods, the historical duration of the historical first moment from the corresponding historical target medical time period, and the first actual number of patients visiting the corresponding historical target medical time period into the first neural network for training. The historical first moment is the moment when the historical symptom information and the historical medical time period on the historical target date are received.

[0083] The data prediction unit is used to input the second patient quantity change characteristics of multiple consecutive historical medical treatment periods within a second time period and equal to the duration of a target medical treatment period for each doctor matching the disease field into a pre-trained second patient quantity prediction model to obtain the second predicted patient quantity output by the second patient quantity prediction model, where the second time period is before the first moment and continuous with the first moment; the second patient quantity prediction model is obtained by inputting the second historical patient quantity change characteristics of multiple consecutive historical medical treatment periods within the historical second time period and equal to the duration of the target medical treatment period for each doctor and the corresponding second actual number of patients as a training sample set into the second neural network for training.

[0084] The second historical time period is before the first historical moment and is continuous with the first historical moment, and the first historical moment is the moment when the historical symptom information and the historical medical treatment period on the historical target date are received.

[0085] The data weighting unit is used to perform weighted summation on the first predicted number of patients and the second number of patients corresponding to each doctor to obtain the final predicted number of patients for each doctor in the target medical treatment period.

[0086] The data acquisition unit is also used to obtain the professional level of each doctor matching the disease field.

[0087] The data acquisition unit is also used to obtain the medical treatment distance between the positioning information of the user terminal and the positioning information of the preset tertiary hospitals to which each doctor belongs.

[0088] The data output unit is used to input the final predicted number of patients for each doctor during the target medical treatment period, the professional level of each doctor matching the disease field, the medical treatment distance between the user terminal's location information and the preset location information of the tertiary hospital to which each doctor belongs, into the pre-trained medical treatment experience scoring model, and output the medical treatment experience score for each doctor.

[0089] Among them, the medical experience score is negatively correlated with the final predicted number of patients, positively correlated with the professional level of doctors matching the disease field, and negatively correlated with the medical distance. Among them, the medical experience scoring model is trained by inputting multiple historical final predicted numbers of patients, the corresponding historical doctor's professional level, and the corresponding historical medical distance as a training sample set into the third neural network.

[0090] A data transmission unit, used to select the top N doctors in the medical experience ranking and transmit them to the user terminal for display;

[0091] The data receiving unit is also used to receive a selected doctor from the top N doctors selected by the user and sent by the user terminal.

[0092] The data sending unit is used to generate medical guidance and triage information based on the location information of the user terminal, the location information of the tertiary hospital to which the selected doctor belongs, and the preset medical treatment process of the selected doctor, and send the medical guidance and triage information to the user terminal for display.

[0093] In a possible implementation, the data receiving unit is used to receive a medical experience score sent by a patient based on a user terminal after the patient selects a doctor for medical treatment;

[0094] A difference determination unit, used to determine the score difference between the received medical experience score and the medical experience score output by the medical experience scoring model;

[0095] The model updating unit is used to update the medical experience scoring model according to the score difference when the score difference is lower than the set difference threshold.

[0096] In a possible implementation, the data acquisition unit is specifically used to obtain the working years, professional title, educational background, and job title of each doctor matching the disease field; to find the first professional score corresponding to the working years, the second professional score corresponding to the professional title, the third professional score corresponding to the educational background, and the fourth professional score corresponding to the job title; to perform a weighted average of the first professional score, the second professional score, the third professional score, and the fourth professional score to obtain a comprehensive professional score, and to use the comprehensive professional score as the professional level of the doctor matching the disease field.

[0097] In one possible implementation, the feature extraction unit is specifically used to extract the first historical dependency of the number of patients visiting the hospital in the target medical treatment period in each of multiple historical days in time series based on the first deep confidence DBN network, and fuse the first historical number of patients visiting the hospital in the target medical treatment period in each of multiple historical days and the first dependency to obtain the first patient number change feature of each doctor matching the disease field in the target medical treatment period.

[0098] In one possible implementation, the feature extraction unit is specifically used to extract the second dependency in time series of the second historical number of patients visiting the medical treatment period for multiple consecutive historical medical treatment periods with the same duration as the target medical treatment period based on the second deep confidence DBN network, and fuse the second historical number of patients visiting the medical treatment period for multiple consecutive historical medical treatment periods with the same duration as the target medical treatment period to obtain the second patient number change characteristics of multiple consecutive historical medical treatment periods with the same duration as the target medical treatment period for each doctor matching the disease field within the second time period.

[0099] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A guidance and triage management method based on smart medical care, characterized in that: Applied to a cloud server, the method comprises: At a first moment, receiving symptom information transmitted by a patient based on a user terminal and a target medical consultation period on a target date; Acquire the disease domain of the patient's disease based on the symptom information, and when it is determined that the disease domain is a seasonal disease, determine a doctor matching the disease domain in a tertiary hospital in the target area; For each doctor matching the disease field, obtain the number of first historical patients visiting the target medical treatment period in a plurality of historical days in a first time period before the first moment, and obtain the number of second historical patients visiting the target medical treatment period in a plurality of consecutive historical medical treatment periods in a second time period before the first moment and having a duration equal to that of the target medical treatment period; Extract the first patient quantity change characteristics of each doctor matching the disease field in the target medical treatment period according to the first historical patient quantity of each target medical treatment period in a plurality of historical days, and extract the second patient quantity change characteristics of each doctor matching the disease field in a plurality of consecutive historical medical treatment periods of equal duration to the target medical treatment period in the second time period according to the second historical patient quantity of each doctor matching the disease field; Input the first patient quantity change characteristics of each doctor matching the disease field in the target medical treatment period and the duration from the first moment to the target medical treatment period into the pre-trained first patient quantity prediction model to obtain the first predicted patient quantity output by the first patient quantity prediction model, wherein the first patient quantity prediction model is obtained by inputting a training sample set consisting of the first historical patient quantity change characteristics of multiple historical target medical treatment periods of each doctor, the historical duration from the historical first moment to the corresponding historical target medical treatment period, and the first actual number of patients visiting the corresponding historical target medical treatment period into the first neural network for training, wherein the historical first moment is the moment when the historical symptom information and the historical medical treatment period on the historical target date are received; Input the second patient quantity change characteristics of a plurality of consecutive historical medical treatment periods of each doctor matching the disease field within a second time period and equal in length to the target medical treatment period into the pre-trained second patient quantity prediction model to obtain the second predicted patient quantity output by the second patient quantity prediction model, wherein the second time period is before the first moment and continuous with the first moment; the second patient quantity prediction model is obtained by inputting the second historical patient quantity change characteristics of a plurality of consecutive historical medical treatment periods of each doctor within a historical second time period and equal in length to the target medical treatment period and the corresponding second actual number of patients as a training sample set into the second neural network for training, wherein the historical second time period is before the historical first moment and continuous with the historical first moment, and the historical first moment is the moment when the historical symptom information and the historical medical treatment period on the historical target date are received; Taking a weighted sum of the first predicted number of patients and the second number of patients corresponding to each doctor, to obtain a final predicted number of patients for each doctor in the target medical treatment period; Get the expertise of each doctor matching the condition area; Obtaining the medical treatment distance between the location information of the user terminal and the location information of the tertiary hospitals to which each doctor belongs; The final predicted number of patients of each doctor in the target medical treatment period, the professional level of each doctor matching the disease field, and the medical treatment distance between the location information of the user terminal and the location information of the preset tertiary hospital to which each doctor belongs are input into the pre-trained medical treatment experience scoring model, and the medical treatment experience score of each doctor is output, wherein the medical treatment experience score is negatively correlated with the final predicted number of patients, positively correlated with the professional level of the doctor matching the disease field, and negatively correlated with the medical treatment distance; wherein the medical treatment experience scoring model is obtained by inputting multiple historical final predicted numbers of patients, corresponding historical doctor professional levels, and corresponding historical medical treatment distances as training sample sets into the third neural network for training; Select the top N doctors in terms of medical experience scores and transmit them to the user terminal for display; Receive a selected doctor from the top N ranked doctors selected by the user sent by the user terminal; Based on the location information of the user terminal, the location information of the tertiary hospital to which the selected doctor belongs, and the preset medical process of the selected doctor, medical guidance and triage information is generated, and the medical guidance and triage information is sent to the user terminal for display.

2. The method according to claim 1, characterized in that After sending the medical guidance and triage information to the user terminal for display, the method includes: receiving a medical experience score sent by the patient based on the user terminal after the patient selects a doctor for medical treatment; Determine a score difference between the received medical experience score and the medical experience score output by the medical experience scoring model; When the score difference is lower than a set difference threshold, the medical experience scoring model is updated according to the score difference.

3. The method according to claim 1, characterized in that The obtaining of the professional level of each doctor matching the disease field includes: Obtain the working years, professional title, education background, and job title of each doctor matching the disease field; Find the first professional score corresponding to the working years, the second professional score corresponding to the professional title, the third professional score corresponding to the academic qualifications, and the fourth professional score corresponding to the job title; The first professional score, the second professional score, the third professional score and the fourth professional score are weighted averaged to obtain a comprehensive professional score, and the comprehensive professional score is used as the professional level of the doctor matching the disease field.

4. The method according to claim 1, characterized in that: The first patient number change characteristics of each doctor matching the disease field in the target medical treatment period are extracted based on the first historical number of patients during the target medical treatment period in the respective multiple historical days, including: According to the first deep confidence DBN network, the first time-series dependency of the first historical number of patients visiting the target medical treatment period in multiple historical days is extracted, and the first historical number of patients visiting the target medical treatment period in multiple historical days and the first dependency are fused to obtain the change characteristics of the first number of patients in the target medical treatment period for each doctor matching the disease field.

5. The method according to claim 1, characterized in that The second historical number of patients in a plurality of consecutive historical medical treatment periods of equal duration to the target medical treatment period is extracted according to the second historical number of patients in a plurality of consecutive historical medical treatment periods of equal duration to the target medical treatment period for each doctor matching the disease field, and the second number of patients in a plurality of consecutive historical medical treatment periods of equal duration to the target medical treatment period within the second time period, including: According to the second deep confidence DBN network, the second dependency in time series of the second historical number of patients visiting the medical treatment in multiple consecutive historical medical treatment periods with the same duration as the target medical treatment period is extracted, and the second historical number of patients visiting the medical treatment in multiple consecutive historical medical treatment periods with the same duration as the target medical treatment period is fused to obtain the second patient number change characteristics of multiple consecutive historical medical treatment periods with the same duration as the target medical treatment period for each doctor matching the disease field within the second time period.

6. A guidance and triage management device based on smart medical care, characterized in that: Configured on a cloud server, the device includes: A data receiving unit, configured to receive, at a first moment, symptom information of a patient transmitted based on a user terminal and a target medical treatment period on a target date; A data acquisition unit, for acquiring the disease field to which the symptom information belongs when it is determined that the disease suffered by the patient is a seasonal disease according to the symptom information, and determining a doctor matching the disease field in a tertiary hospital in a target area; The data acquisition unit is further configured to acquire, for each doctor matching the disease field, a first historical number of patients visiting the target medical treatment period in a plurality of historical days in a first time period before the first moment, and acquire a second historical number of patients visiting the target medical treatment period in a plurality of consecutive historical medical treatment periods in a second time period before the first moment and having a duration equal to that of the target medical treatment period; A feature extraction unit is used to extract the first patient quantity change characteristics of each doctor matching the disease field in the target medical treatment period according to the first historical patient quantity of each target medical treatment period in a plurality of historical days, and to extract the second patient quantity change characteristics of each doctor matching the disease field in a plurality of consecutive historical medical treatment periods with a duration equal to that of the target medical treatment period within the second time period according to the second historical patient quantity of each doctor matching the disease field; a data prediction unit, for inputting the first patient quantity change characteristics of each doctor matching the disease field in the target medical treatment period and the duration from the first moment to the target medical treatment period into a pre-trained first patient quantity prediction model, so as to obtain a first predicted patient quantity output by the first patient quantity prediction model, wherein the first patient quantity prediction model is obtained by inputting a training sample set consisting of the first historical patient quantity change characteristics of multiple historical target medical treatment periods of each doctor, the historical duration from the historical first moment to the corresponding historical target medical treatment period, and the first actual number of patients visiting the corresponding historical target medical treatment period into a first neural network for training, wherein the historical first moment is the moment of receiving historical symptom information and the historical medical treatment period on the historical target date; The data prediction unit is used to input the second patient quantity change characteristics of multiple consecutive historical medical treatment periods of each doctor matching the disease field within a second time period and equal to the duration of the target medical treatment period into the pre-trained second patient quantity prediction model to obtain the second predicted patient quantity output by the second patient quantity prediction model, wherein the second time period is before the first moment and continuous with the first moment; the second patient quantity prediction model is obtained by inputting the second historical patient quantity change characteristics of multiple consecutive historical medical treatment periods of each doctor within the historical second time period and equal to the duration of the target medical treatment period and the corresponding second actual number of patients as a training sample set into the second neural network for training, wherein the historical second time period is before the historical first moment and continuous with the historical first moment, and the historical first moment is the moment when the historical symptom information and the historical medical treatment period on the historical target date are received; a data weighting unit, configured to perform a weighted summation of the first predicted number of patients and the second number of patients corresponding to each doctor, so as to obtain a final predicted number of patients for each doctor in the target medical treatment period; The data acquisition unit is further used to acquire the professional level of each doctor matching the disease field; The data acquisition unit is further used to acquire the medical treatment distance between the location information of the user terminal and the location information of the preset tertiary hospitals to which each doctor belongs; A data output unit is used to input the final predicted number of patients of each doctor in the target medical treatment period, the professional level of each doctor matching the disease field, and the medical treatment distance between the location information of the user terminal and the location information of the preset tertiary hospital to which each doctor belongs, into a pre-trained medical treatment experience scoring model, and output the medical treatment experience score of each doctor, wherein the medical treatment experience score is negatively correlated with the final predicted number of patients, positively correlated with the professional level of the doctor matching the disease field, and negatively correlated with the medical treatment distance; wherein the medical treatment experience scoring model is obtained by inputting a plurality of historical final predicted numbers of patients, corresponding historical professional levels of doctors, and corresponding historical medical treatment distances as a training sample set into a third neural network for training; A data transmission unit, used to select the top N doctors in the medical experience ranking and transmit them to the user terminal for display; The data receiving unit is further used to receive a selected doctor from the top N doctors selected by a user from a user terminal; A data sending unit is used to generate medical guidance and triage information based on the location information of the user terminal, the location information of the tertiary hospital to which the selected doctor belongs, and the preset medical treatment process of the selected doctor, and send the medical guidance and triage information to the user terminal for display.

7. The device according to claim 6, characterized in that The data receiving unit is used to receive the medical experience score sent by the patient based on the user terminal after the patient selects a doctor for medical treatment; a difference determination unit, configured to determine a score difference between the received medical experience score and the medical experience score output by the medical experience scoring model; A model updating unit is used to update the medical experience scoring model according to the scoring difference when the scoring difference is lower than a set difference threshold.

8. The device according to claim 6, characterized in that The data acquisition unit is specifically used to obtain the working years, professional title, academic qualifications, and job title of each doctor matching the disease field; find the first professional score corresponding to the working years, the second professional score corresponding to the professional title, the third professional score corresponding to the academic qualifications, and the fourth professional score corresponding to the job title; perform weighted average of the first professional score, the second professional score, the third professional score, and the fourth professional score to obtain a comprehensive professional score, and use the comprehensive professional score as the professional level of the doctor matching the disease field.

9. The device according to claim 6, characterized in that The feature extraction unit is specifically used to extract the first time-series dependency of the first historical number of patients visiting the target medical treatment period in multiple historical days respectively according to the first deep confidence DBN network, and fuse the first historical number of patients visiting the target medical treatment period in multiple historical days respectively and the first dependency to obtain the first patient number change feature of each doctor matching the disease field in the target medical treatment period.

10. The device according to claim 6, characterized in that The feature extraction unit is specifically used to extract the second time-series dependency of the second historical patient numbers of multiple consecutive historical medical treatment periods with the same duration as the target medical treatment period based on the second deep confidence DBN network, and fuse the second historical patient numbers of multiple consecutive historical medical treatment periods with the same duration as the target medical treatment period to obtain the second patient number change characteristics of multiple consecutive historical medical treatment periods with the same duration as the target medical treatment period for each doctor matching the disease field within the second time period.