A method, system, and device for matching physician resources based on vital sign data.

Abstract

This application relates to the field of medical data information technology, specifically disclosing a method, system, and device for matching doctor resources based on vital sign data. The method mainly includes receiving patient data sent from a patient terminal, including the patient's vital sign data; inputting the patient data into a deep learning model to obtain patient assessment data, including a disease type probability distribution; acquiring doctor information from online doctors; matching the patient's disease type probability distribution with the doctor information to obtain the total matching degree for each online doctor; and selecting the doctors with the highest total matching degree as the initial matching result. This application, by uploading disease and vital sign data to the backend, enables automatic doctor matching and remote treatment intervention, facilitating remote communication and treatment between doctors and patients, improving the efficiency of medical resource utilization, shortening patient waiting time, and ensuring patients receive timely professional medical services.

Description

Technical Field

[0001] This application relates to the field of medical data information technology, and in particular to a method for matching doctor resources based on vital sign data, a system for matching doctor resources based on vital sign data, and a device for matching doctor resources based on vital sign data. Background Technology

[0002] Regular checkups, maintaining healthy lifestyle and dietary habits, and proactively addressing physical discomfort are all signs of prioritizing one's health. Seeking timely medical attention means actively seeking medical help when abnormalities or symptoms are detected, without delaying treatment. This health awareness helps in the early detection and treatment of diseases, reducing the risk of disease progression and improving treatment outcomes.

[0003] In traditional healthcare systems, patients typically need to self-diagnose their symptoms and attempt to find a specialist doctor. However, this approach presents several challenges. First, patients often lack medical expertise and clinical experience, increasing the risk of misdiagnosis and incorrect doctor matching and treatment plans. Second, patients expend significant time and energy searching for and screening doctors, potentially delaying treatment, especially in emergencies. Furthermore, self-selection of doctors can lead to inaccurate matching, as patients may not have a comprehensive understanding of all doctors' areas of expertise and experience. Therefore, self-diagnosis and self-selection of doctors suffer from drawbacks such as low diagnostic accuracy, high time and energy consumption, and inaccurate doctor matching. Summary of the Invention

[0004] This application provides a method, system, and device for matching doctors based on vital sign data. This solves the technical problems in the prior art, such as low diagnostic accuracy, high time and energy consumption, and inaccurate doctor matching, which often require patients to assess their own symptoms in order to find a doctor of appropriate specialty. It can improve the accuracy of medical diagnosis, optimize the doctor matching process, and accelerate the doctor response time, so as to better meet the needs of patients for medical services.

[0005] Firstly, this application provides a doctor resource matching method based on vital sign data, which includes the following steps: receiving patient data sent by a patient terminal, the patient data including the patient's vital sign data; inputting the patient data into a deep learning model to obtain patient evaluation data, the evaluation data including disease type probability distribution. ; Obtain doctor-side information from the online doctor's end, and calculate the probability distribution of the patient's disease type. The system matches the doctor's information to obtain the total matching score for each online doctor. The doctor with the highest total matching score is selected as the preliminary matching result. A diagnosis request is sent to the doctor based on the preliminary matching result.

[0006] Furthermore, it receives response information returned by the doctor based on the diagnosis request.

[0007] Furthermore, the doctor's information includes their professional field. and interdisciplinary fields, In the formula, Represents the m-th professional field; , Represents the nth disease type The probability of the patient's disease type; the probability distribution of the disease type. Matching the doctor's information to obtain the total matching degree for each online doctor specifically includes: calculating the total matching degree for each doctor. and the types of diseases the patient may have Matching degree between , for: In the formula, This indicates that the patient has a disease. The probability, Representative doctor In the professional field Professional ability score Representative doctor In disease type The score for cross-disciplinary collaboration.

[0008] Furthermore, the patient data also includes disease-related features, and the doctor's information includes professional level; the disease-related features are input into the trained support vector model to obtain the complexity of the disease;

[0009] Based on the initial matching results and the complexity level of the disease, doctors of the corresponding professional level are matched.

[0010] Furthermore, the complexity includes low-complexity, medium-complexity, and high-complexity diseases, and the professional level includes primary, intermediate, and advanced professional levels. For low-complexity, medium-complexity, and high-complexity diseases, primary, intermediate, and advanced professional doctors are preferentially matched from the preliminary matching results, respectively, and diagnostic requests are sent. If no response is received within a specified time, doctors of other professional levels are added.

[0011] Furthermore, the assessment data also includes the degree of urgency, and the doctor's information also includes emergency level information, which includes level A. After obtaining the preliminary matching results, doctors with level A are matched first from the preliminary matching results, and other doctors are matched after a time threshold T has elapsed.

[0012] Furthermore, T is: In the formula, T represents the initial time threshold, T represents the actual time threshold, and U represents the urgency level. , k represents an adjustable parameter used to control the degree to which the urgency level affects the time threshold.

[0013] Furthermore, the steps to obtain the overall score include:

[0014] The response time score is calculated based on the doctor's historical average response time. Specifically, the response time of the doctor's most recent N diagnostic requests is normalized and then mapped to a 0-10 scale.

[0015] The treatment effectiveness score is obtained by weighting the patient recovery rate and satisfaction score. The patient recovery rate is the proportion of cases treated by the doctor whose physiological indicators returned to normal, and the satisfaction score is the average score of patients' evaluations of the medical services. A comprehensive score is then obtained. , for:

[0016] ;

[0017] In the formula, Represents the response time score. The treatment effect score represents the treatment effect score. This is a weighting coefficient for treatment effectiveness;

[0018] If the overall score If the value is greater than the preset value, the emergency response level for the doctor will be Level A.

[0019] Secondly, this application provides a doctor resource matching system based on vital sign data, which adopts the doctor resource matching method based on vital sign data as described in the first aspect, and includes: a receiving module, an evaluation module, a matching module, and a sending module.

[0020] The receiving module receives patient data sent from the patient terminal, including the patient's vital signs data. The evaluation module inputs the patient data into a deep learning model to obtain the patient's evaluation data, which includes a disease type probability distribution. The matching module is used to obtain doctor-side information from online doctors and to determine the probability distribution of the patient's disease type. The system matches the doctor's information to obtain the total matching score for each online doctor. The doctor with the highest total matching score is selected as the preliminary matching result. The sending module is used to send a diagnosis request to the doctor based on the preliminary matching result.

[0021] Thirdly, this application provides a physician resource matching device based on vital sign data, which includes a memory and a processor.

[0022] The memory is used to store computer programs; the processor is used to execute the computer programs to implement the steps of the physician resource matching method based on vital sign data as described in the first aspect.

[0023] The technical solution provided in this application has at least the following technical effects or advantages:

[0024] By uploading patient symptoms data to the backend, the system automatically matches patients with doctors and provides remote treatment intervention. This effectively solves the technical problems of low diagnostic accuracy, high time and energy consumption, and inaccurate doctor matching that often arise when patients need to assess their own symptoms to find a suitable doctor. It enables remote communication and treatment between doctors and patients, improving the efficiency of medical resource utilization, shortening patient waiting times, and ensuring timely access to professional medical services. This approach also allows for rapid response in emergencies, reducing the risk of condition deterioration due to waiting for treatment, and improving the quality and efficiency of medical services. Attached Figure Description

[0025] Figure 1 This is a flowchart illustrating the physician resource matching method based on vital sign data in this application;

[0026] Figure 2 This is a flowchart illustrating the process for matching Grade A doctors in this application. Detailed Implementation

[0027] To address the technical challenges of low diagnostic accuracy, high time and energy consumption, and inaccurate doctor matching caused by patients typically having to self-assess their symptoms in order to find a suitable doctor, this application provides a doctor resource matching method based on vital sign data. This method can improve the accuracy of medical diagnosis, optimize the doctor matching process, and accelerate doctor response time, thereby better meeting patients' needs for medical services.

[0028] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0029] like Figures 1-2As shown, this embodiment provides a doctor resource matching method based on vital sign data, which includes the following steps: receiving patient data sent by the patient terminal, the patient data including the patient's vital sign data; inputting the patient data into a deep learning model to obtain the patient's assessment data, the assessment data including the disease type probability distribution. ; Obtain doctor-side information from online doctors and analyze the probability distribution of patients' disease types. The system matches the information with the doctor's information to obtain the total matching score for each online doctor, and selects the doctor with the highest total matching score as the preliminary matching result; based on the preliminary matching result, it sends a diagnosis request to the doctor's end; and it receives the response information returned by the doctor's end based on the diagnosis request.

[0030] Conditions may require the use of predictive models and matching mechanisms to allocate appropriate levels of medical resources, such as cases of sudden cardiac arrest, stroke, cancer treatment, and allergic reactions. Vital signs monitoring devices can monitor patients' vital signs in real time, such as heart rate, blood pressure, and blood oxygen saturation. Analyzing these vital signs data can provide a preliminary assessment of the patient's symptoms to match them with a suitable doctor. Furthermore, a comprehensive assessment can be made by combining the patient's medical history and symptom descriptions.

[0031] To build a disease diagnosis model, machine learning models, such as deep learning models or rule-based expert systems, can be used. The input of the model is preprocessed vital sign data, and the output is the symptoms or disease types that the patient may have.

[0032] Taking a deep learning-based convolutional neural network (CNN) as an example of a machine learning model for disease diagnosis, after collecting a large amount of labeled vital sign data, the data is preprocessed, such as signal filtering and standardization, before a disease diagnosis model is constructed. A suitable CNN model for disease diagnosis is designed, including convolutional layers, pooling layers, and fully connected layers. The CNN model can be represented as: In the formula, X represents the input vital sign data. and These represent the weight matrices of the two fully connected layers, and These represent the biases of the two fully connected layers, respectively. Represents the activation function. Output for multi-class classification problems.

[0033] Doctor-side information includes professional fields and interdisciplinary fields, In the formula, Representing the m-th professional field, , Represents the nth disease type The probability of the patient's disease type; the probability distribution of the patient's disease type. Matching the information with the doctor's end to obtain the total matching degree for each online doctor specifically includes: calculating the total matching degree for each doctor. and the types of diseases the patient may have Matching degree between , for: In the formula, This indicates that the patient has a disease. The probability, Representative doctor In the professional field Professional ability score Representative doctor In disease type The score for cross-disciplinary collaboration.

[0034] To measure the difference between the model output and the real symptom label, an appropriate loss function is selected. Since each patient only needs to be matched with a doctor in a single specialty field at a time, that is, each patient's main symptom corresponds to a unique disease type, the multi-class cross-entropy loss function is adopted.

[0035] If there are n possible disease types, the model output is an n-dimensional probability distribution vector. , , representing the model's predicted probability for each disease type. For a sample, the true disease type label is represented as an n-dimensional encoded vector y. In this context, only one element is 1, representing the true disease type. Therefore, the loss function can be expressed as: In the formula, , i represents the number of groups. Representative model output The cross-entropy loss between the model's output probability vector and the true label. By minimizing the multi-class cross-entropy loss function, the model's output probability distribution vector can be optimized. The goal is to approximate the true disease type label y as closely as possible. Since the severity and impact of different disease types may vary, the loss function can be adjusted according to specific needs, such as using a weighted cross-entropy loss function, to better reflect the importance of different disease types. After deriving the predicted probability distribution of the possible disease types a patient may have, information from a pre-established doctor database is matched with online doctor information to automatically select a matching doctor.

[0036] In this embodiment, the matching degree of doctors is evaluated by considering multiple factors such as the probability distribution of patient disease types, doctors' professional fields, and collaboration in interdisciplinary fields, thereby achieving preliminary matching. For example, the doctor's professional field is... , For every doctor and the types of diseases the patient may have The matching degree between them can be calculated using the formula: In the formula, This indicates that the patient has a disease. The probability, Representative doctor In the professional field Professional ability score Representative doctor In disease type The score for cross-disciplinary collaboration, if the doctor is involved in disease type Having experience collaborating in other relevant fields can provide additional points for better matching. and These represent weighting coefficients used to balance the matching degree between the patient's disease type probability distribution and the doctor's specialty, as well as the matching degree between cross-disciplinary collaboration. The final matching degree matrix can be represented as a... The matrix is ​​given, where m represents the number of doctors and n represents the number of possible disease types of patients. After calculating the matching degree matrix according to the matching degree evaluation formula, the matching degree of various disease types can be comprehensively considered to calculate the total matching degree for each doctor. The doctors with the highest matching degree are selected as the preliminary matching results, and the doctors' responses are awaited.

[0037] By uploading patients' vital signs data to the backend, the system automatically matches patients with doctors and enables remote treatment intervention. This remote communication and treatment between doctors and patients improves the efficiency of medical resource utilization, shortens patient waiting times, and ensures patients receive timely professional medical services. This approach also allows for rapid response in emergencies, reducing the risk of condition deterioration due to waiting for treatment and improving the quality and efficiency of medical services.

[0038] Patient data also includes disease-related features, and doctor-side information includes professional level. Disease-related features are input into a trained support vector model to obtain the disease complexity. Based on the preliminary matching results and the disease complexity level, doctors of corresponding professional levels are matched. Complexity includes low-complexity, medium-complexity, and high-complexity diseases, and professional levels include junior, intermediate, and senior professionals. For low-complexity, medium-complexity, and high-complexity diseases, doctors of junior, intermediate, and senior professional levels are preferentially matched from the preliminary matching results, respectively, and diagnostic requests are sent. If no response is received within a specified time, doctors of other professional levels are added.

[0039] The process for obtaining a professional level includes:

[0040] Four basic indicators were extracted from the doctor database: educational background, professional certificates, years of clinical experience, and scientific research achievements.

[0041] Each indicator is standardized, and a weighted average is used to obtain the scoring range.

[0042] Professional levels are categorized by scoring range:

[0043] Elementary level: S < 60 points;

[0044] Intermediate: 60 ≤ S < 85 points;

[0045] Advanced: S ≥ 85 points.

[0046] The aforementioned matching method may result in complex cases being assigned to doctors with lower levels of expertise, potentially impacting the quality of medical services and patient outcomes. Conversely, senior doctors may handle simpler cases, wasting their expertise and time, reducing the efficiency of medical resource utilization, and possibly leading to patient dissatisfaction, thus affecting the hospital's reputation and patient trust. In this embodiment, based on the above, support vector machine analysis of patient vital sign data and symptoms allows for case assessment and determination of complexity levels. Doctors are then categorized according to their expertise levels, ensuring that each case receives services from a doctor of the appropriate level.

[0047] When training a support vector model, features related to the patient's condition are used as input. These features include, but are not limited to, physiological indicators, textual descriptions of symptoms, and basic patient information. Physiological indicators include blood pressure, heart rate, body temperature, and blood test results. Symptom descriptions include the patient's described symptoms, pain intensity, and duration. Basic patient information may include age, gender, and past medical history. These features can be numerical, textual, or even categorical, and are processed and feature-engineered according to the specific situation. The output is a corresponding complexity label, for example, -1 represents low complexity, 0 represents medium complexity, and 1 represents high complexity. After training the model, the complexity level of the disease is evaluated based on the input features.

[0048] Doctors can be assessed on their professional level based on multiple indicators, such as educational background and professional qualifications, clinical experience and technical level, patient satisfaction and medical quality, and professional development and academic contributions. Educational background and professional qualifications include a doctor's academic qualifications, professional qualification certificates, and continuing education. Clinical experience and technical level refer to a doctor's clinical work experience, technical level, and research achievements. Patient satisfaction and medical quality include patient satisfaction, medical quality evaluation, and medical malpractice records. Professional development and academic contributions can include a doctor's academic research achievements, academic exchange activities, and contributions to the medical profession. By weighting and summing the different indicators, a doctor's assessment score can be obtained. By setting different score ranges, doctors' professional levels can be divided into primary, intermediate, and advanced levels from low to high.

[0049] Among the initially matched doctors, those at the appropriate specialty level can be matched based on the complexity of the illness. For low-complexity illnesses, priority is given to matching with doctors at the junior specialty level and sending a diagnosis request to them. If no doctor responds within a specified time, a doctor at the intermediate specialty level is matched and a diagnosis request is sent. If there is still no response within the specified time, a doctor at the senior specialty level is matched last. For medium-complexity illnesses, priority is given to matching with doctors at the intermediate specialty level, followed by doctors at the senior specialty level. If a doctor at the senior specialty level does not respond, a doctor at the junior specialty level is matched last. For high-complexity illnesses, priority is given to matching with doctors at the senior specialty level, followed by doctors at the intermediate specialty level, and finally doctors at the junior specialty level. Categorizing doctors by specialty level helps to better match patients' needs, rationally allocate medical resources, and improve the efficiency and quality of medical services.

[0050] The assessment data also includes the degree of urgency, and the doctor's information includes emergency level information, which includes level A. After obtaining the initial matching results, doctors with level A are matched first. After a time threshold T has elapsed, other doctors are matched.

[0051] In emergency situations such as stroke, sudden heart attack, severe trauma, respiratory distress, and severe bleeding, urgent medical intervention is required to ensure patients receive timely and appropriate medical services. When using deep learning-based convolutional neural networks to analyze vital sign data for preliminary assessment of a patient's symptoms, an additional output layer can be added to the model to evaluate and predict the urgency of the patient's condition. This additional output layer can be designed as a regression layer, outputting a value between 0 and 1, representing the urgency level of the patient, indicating the probability or degree that the patient needs urgent medical intervention. Training the deep learning model requires a training dataset containing vital sign data and corresponding urgency level labels. These labels can be urgency assessments given by doctors based on the patient's condition, or labels assigned based on factors such as the severity of the patient's illness. A fully connected layer is added after the last layer of the convolutional neural network as an additional output layer, outputting a numerical value representing the urgency level. Simultaneously, an appropriate loss function, such as mean squared error, needs to be defined to measure the difference between the predicted percentage of urgency and the true label.

[0052] After inputting a patient's vital sign data, the model not only identifies the patient's possible symptoms or disease types but also provides an assessment of the urgency level as a percentage. This allows for the matching of physician resources based on the patient's level of urgency. When evaluating physician matching based on the probability distribution of patient disease types and collaboration within physicians' areas of expertise and interdisciplinary collaborations, a focus is placed on physicians capable of addressing emergency medical needs. This enables rapid and accurate physician matching in emergency situations, improving patient survival rates and treatment outcomes.

[0053] The steps to obtain a comprehensive score include:

[0054] The response time score is calculated based on the doctor's historical average response time. Specifically, the response time of the doctor's most recent N diagnostic requests is normalized and then mapped to a 0-10 scale.

[0055] Treatment effectiveness score is obtained by weighting patient recovery rate and satisfaction score. Patient recovery rate is the proportion of patients whose physiological indicators return to normal in cases treated by the doctor, and satisfaction score is the average score of patients' evaluation of the medical services; a comprehensive score is then obtained. , for:

[0056] ;

[0057] In the formula, Represents the response time score. The treatment effect score represents the treatment effect score. This is a weighting coefficient for treatment effectiveness;

[0058] If the overall score If the value is greater than the preset value, the emergency response level for the doctor will be Level A.

[0059] Specifically, information such as response time and treatment effectiveness can be incorporated into doctor matching. During the interaction between the matched doctor and patient, the doctor's response time, the treatment plan provided, and its effects are collected. Timing begins when the matched doctor receives the alert and begins handling the emergency, recording the doctor's response time—the time from receiving the response request to providing medical support. After handling the emergency, the doctor records the patient's treatment outcomes, including changes in vital signs and symptom relief, to assess the doctor's treatment effectiveness.

[0060] When recording response time, you can start timing from the time the matched doctor responds to the diagnosis, record the timestamp, start the matching doctor to provide medical support until the emergency is properly handled, stop timing after the matching doctor provides medical support, record the timestamp of the end time, and use the end time to subtract the start time to calculate the matching doctor's response time.

[0061] When recording treatment effects, the collected treatment outcome data can be analyzed to evaluate the doctor's treatment effectiveness, such as whether vital signs are effectively controlled and symptoms are relieved. The evaluated treatment effects are recorded in the patient's medical records or a dedicated treatment effect database. At the same time, feedback from patients or their families can be collected to understand their evaluation and opinions on the treatment provided by the doctor. Taking into account changes in vital signs, symptom relief, and patient feedback, a comprehensive evaluation of the doctor's treatment effectiveness can be conducted.

[0062] T is: In the formula, T represents the initial time threshold, T represents the actual time threshold, and U represents the urgency level. , k represents an adjustable parameter used to control the degree to which the urgency level affects the time threshold.

[0063] By evaluating the doctor's response time and treatment effectiveness, a comprehensive score can be obtained for the doctor's emergency treatment. , In the formula, Represents the response time score. The treatment effectiveness score is determined by a preset threshold and based on the overall score. Doctors are divided into several levels. Those with high comprehensive scores can be classified as level A doctors. After obtaining the initial matching results, the evaluation results of the patient's disease type and urgency percentage are obtained. Level A doctors are matched first from the initial matching doctors. After a time threshold T, level B doctors are matched from the initial matching doctors at the same time.

[0064] The time threshold can be determined based on the patient's urgency. Specifically, the formula for calculating the time threshold T is: In the formula, T represents the initial time threshold, T represents the actual time threshold, and k represents an adjustable parameter used to control the degree of influence of urgency on the time threshold, representing the urgency level, with a value between 0 and 1.

[0065] Prioritizing matching with doctors significantly improves the effective utilization of medical resources and the efficiency of patient care. While ensuring the quality of medical services, the professional competence and experience of prioritized doctors better meet patient needs, directly improving the overall quality of medical services and patient satisfaction. The system incorporates the patient's urgency as an adjustment factor in the matching time, demonstrating strong flexibility and adaptability. In urgent patient situations, the system automatically shortens the time for switching between all doctors, ensuring that patients receive appropriate medical services in the shortest possible time. This plays a crucial role in improving emergency response efficiency and reducing the risks patients may face while waiting for treatment.

[0066] Example 2

[0067] This embodiment provides a doctor resource matching system based on vital sign data, which adopts the doctor resource matching method based on vital sign data as in Embodiment 1, and includes: a receiving module, an evaluation module, a matching module, and a sending module.

[0068] The receiving module receives patient data sent from the patient terminal, including the patient's vital signs. The evaluation module inputs the patient data into a deep learning model to obtain evaluation data, which includes the probability distribution of disease types. The matching module is used to obtain doctor-side information from online doctors and to determine the probability distribution of patients' disease types. The system matches the information with the doctor's information to obtain the total matching score for each online doctor. The doctor with the highest total matching score is selected as the initial matching result. The sending module is used to send a diagnosis request to the doctor based on the initial matching result.

[0069] This physician resource matching system based on vital sign data has all the advantages of the vital sign data-based physician resource matching method and can automatically implement all the steps of the vital sign data-based physician resource matching method.

[0070] Example 3

[0071] This application provides a physician resource matching device based on vital sign data, which includes a memory and a processor.

[0072] The memory is used to store computer programs; the processor is used to execute the computer programs to implement the steps of the doctor resource matching method based on vital sign data as in Embodiment 1.

[0073] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0074] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the functions specified in one or more boxes. Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the invention.

Claims

1. A physician resource matching method based on vital sign data, characterized in that, It comprises the following steps: Receiving patient data sent by the patient end, the patient data comprising vital sign data of the patient; inputting the patient data into a deep learning model to obtain evaluation data of the patient, the evaluation data comprising a probability distribution of disease types ; Obtaining the doctor side information of the online doctor side, obtaining the disease type probability distribution of the patient Matching the doctor side information, obtaining the total matching degree of each online doctor side, and selecting the top n doctors with the highest total matching degree as the preliminary matching result Sending a diagnosis request to the doctor end based on the preliminary matching result; The evaluation data further comprises emergency degree, and the doctor end information further comprises emergency level information, the emergency level information comprising A level; After obtaining the preliminary matching result, the doctor end with A level is matched preferentially from the preliminary matching result, and after a time threshold T, other doctor ends are matched; The step of obtaining the comprehensive score comprises: The response time score is obtained by calculating the average response time of the doctor in history, specifically: the response time of the doctor in the last N diagnosis requests is normalized and mapped to 0-10 points; The treatment effect score is obtained by weighting the patient recovery rate and the satisfaction score, the patient recovery rate being the proportion of cases treated by the physician in which the physiological indicators returned to normal, and the satisfaction score being the average score given by the patients to the medical treatment service; obtaining a global score , is: ； wherein represents the response time score, represents the treatment effect score, is the treatment effect weight coefficient; If the overall score If the value is greater than the preset value, the emergency response level for the doctor will be Level A.

2. The physician resource matching method based on vital signs data as claimed in claim 1, wherein, Receiving response information returned by the doctor end according to the diagnosis request.

3. The physician resource matching method based on vital signs data as claimed in claim 1, wherein, The doctor-side information includes professional fields and cross fields, wherein, represents the mth professional field; , probability of representing the nth disease type ; determining a probability distribution of disease types for the patient matching the doctor-side information to obtain a total matching degree of each online doctor-side specifically comprising: The matching degree between each doctor and the type of disease the patient can have is calculated as: :​​​ ； wherein, P (Disease) represents the probability that a patient has a disease , P (Professional) represents the professional ability score of a physician in a professional field , P (Cross) represents the cross-field collaboration score of a physician in a disease type , and respectively represent the weight coefficients for balancing the matching degree of the probability distribution of the disease type of a patient and the professional field of a physician and the matching degree of cross-field collaboration.

4. The method of claim 1, wherein, The patient data further comprises disease-related characteristics, and the doctor end information further comprises professional level; the disease-related characteristics are input into the trained support vector model to obtain the complexity of the disease; The doctor with corresponding professional level is matched based on the preliminary matching result and the complexity level of the disease.

5. The physician resource matching method based on vital signs data as claimed in claim 4, wherein, The complexity comprises low complexity disease, medium complexity disease and high complexity disease, and the professional level comprises primary professional, intermediate professional and advanced professional; The low complexity disease, the medium complexity disease and the high complexity disease are preferentially matched with the doctor end of the primary professional, the intermediate professional and the advanced professional respectively and the diagnosis request is sent, and if the response information is not received within a specified time, the doctor end of other professional levels is added.

6. The vital sign data based physician resource matching method as claimed in claim 1, wherein, T is: ； wherein represents the initial set time threshold, T represents the actual time threshold, U represents the urgency level, and k represents an adjustable parameter used to control the degree of influence of the urgency level on the time threshold.

7. A physician resource matching system based on vital sign data, which adopts the physician resource matching method based on vital sign data according to any one of claims 1-6, characterized in that, It comprises: A receiving module for receiving patient data sent by the patient end, the patient data comprising vital sign data of the patient; an evaluation module for inputting the patient data into a deep learning model to obtain evaluation data of the patient, the evaluation data comprising a probability distribution of disease types ; The matching module is used to obtain doctor-side information from online doctors and to perform probability distribution of the patient's disease type. The total matching degree of each online doctor is obtained by matching the doctor's information, and the doctors with the highest total matching degree are selected as the preliminary matching results. A sending module for sending a diagnosis request to the doctor end based on the preliminary matching result.

8. A physician resource matching device based on vital sign data, the device comprising: It comprises: A memory for storing a computer program; A processor for executing the computer program to realize the steps of the doctor resource matching method based on vital sign data according to any one of claims 1-6.

Images