Decision support method, system and device based on vital sign data

Abstract

The invention relates to the technical field of medical information, and particularly discloses a decision support method, system and device based on vital sign data. The method mainly comprises the following steps: receiving patient data sent by a patient end, wherein the patient data comprises vital sign data; inputting patient data into the prediction model to obtain a disease analysis report of the patient; matching the disease analysis report of the patient with doctor information to obtain a matched doctor end, and sending a diagnosis request to the matched doctor end; and receiving response information returned by the matched doctor based on the diagnosis request, and sending the disease analysis report to a doctor terminal of the returned response information to provide decision support. According to the method, the time of manually sorting and analyzing the illness state information by doctors is shortened, and the doctors are helped to better understand the illness state of the patient and formulate a more effective treatment scheme, so that a decision can be made more quickly.

Description

Technical Field

[0001] The present application relates to the field of medical information technology, and in particular to a decision support method based on vital signs data, a decision support system based on vital signs data, and a decision support device based on vital signs data. Background Art

[0002] Online medical consultation means that patients can use tablet devices for remote medical consultation and diagnosis and treatment. Patients can make video calls or online chats with doctors through tablet devices, describe their symptoms to doctors and receive doctors' diagnoses and suggestions. Doctors can also observe patients' symptoms, provide remote monitoring and prescribe drugs through tablet devices. This method allows patients to communicate with doctors at home or any convenient place, avoiding the time and transportation costs of going to medical institutions, and also improving the efficiency of medical resource utilization.

[0003] However, since patients cannot have face-to-face consultations, it is difficult for doctors to obtain sufficient clinical information and intuitive perception. In addition, they also have difficulty obtaining complete medical records and test results, which makes it difficult for doctors to quickly judge patients' symptoms, which may lead to delayed diagnosis and treatment, thereby prolonging the patient's condition, affecting the treatment effect, and even leading to serious consequences. Therefore, although online consultations provide convenience, doctors lack sufficient clinical information and intuitive perception, which may affect their ability to provide decision support, thereby affecting treatment outcomes and patient satisfaction. Summary of the invention

[0004] The present application solves the problem in the prior art that it is difficult for doctors to quickly judge the symptoms of patients during online medical consultations by providing a decision support method, system and device based on vital signs data. By directly accepting the vital signs data sent by the patient and using a predictive model to quickly generate a disease analysis report and send it to the doctor, decision support is provided to the doctor, which is conducive to the doctor's rapid judgment of the patient's symptoms, improves the diagnosis efficiency, and shortens the time patients wait for diagnosis results. This efficiency improvement is particularly critical for the timely treatment of critically ill patients.

[0005] The present application provides a decision support method based on vital signs data, which includes the following steps: receiving patient data sent by a patient end, the patient data including vital signs data; inputting the patient data into a prediction model to obtain a disease analysis report of the patient; obtaining doctor information from a doctor end; matching the disease analysis report of the patient with the doctor information to obtain a matched doctor end, and sending a diagnosis request to the matched doctor end; receiving response information returned by the matched doctor based on the diagnosis request, and sending the disease analysis report to the doctor end of the returned response information, thereby providing decision support to the doctor end.

[0006] By directly receiving vital signs data sent by the patient and using predictive models to quickly generate disease analysis reports and send them to the doctor, decision support is provided to the doctor, reducing the time doctors spend manually organizing and analyzing medical information, helping them better understand the patient's condition, and helping doctors quickly judge the patient's symptoms and develop more effective treatment plans, thereby enabling them to make decisions faster and improving diagnostic efficiency.

[0007] Furthermore, the disease analysis report includes the disease type and the time required for consultation T1; inputting patient data into the prediction model specifically includes: inputting patient data into a multi-class logistic regression model to obtain the disease type; inputting patient data into the regression model to obtain the time required for consultation T1.

[0008] Furthermore, the doctor information includes a diagnosis efficiency DE, and the diagnosis efficiency DE is: Where, DI i Represents the evaluation effect of the doctor's i-th diagnosis DI, DT i represents the time taken by the doctor for the i-th diagnosis, and n represents the number of diagnoses made by the doctor.

[0009] Furthermore, the doctor information also includes the doctor's professional field, professional level and status information, and the status information includes the remaining time for consultation and idle status; matching the patient's disease analysis report with the doctor information to obtain the matched doctor end specifically includes: obtaining the doctor end in idle status and with the remaining time for consultation less than the preset time T2 and constructing a doctor end set D1; matching the disease type with the doctor's professional field in the doctor end set D1 to obtain a doctor end set D2 that matches the disease type; obtaining the number N1 of doctor ends in the doctor end set D2, and determining whether the number N1 of doctor ends is greater than the preset number N2, and if so, executing matching mechanism one, otherwise executing matching mechanism two.

[0010] Furthermore, the matching mechanism 1 includes: constructing a doctor end set d1, putting a preset number of doctor ends with the highest professional level in the doctor end set D2 into the doctor end set d1, and taking the remaining doctor ends in the doctor end set D2 as the doctor end set d2; sending a diagnosis request to all doctor ends in the doctor end set d1, and if no response information based on the diagnosis request is received within the preset time T2, sending a diagnosis request to the doctor end set d2.

[0011] Further, the second matching mechanism includes: constructing a doctor terminal set d3, putting a preset number of doctor terminals with the highest diagnosis efficiency DE in the doctor terminal set D2 into the doctor terminal set d3, and regarding the remaining doctor terminals in the doctor terminal set D2 as the doctor terminal set d4; determining whether the required consultation time T1 is greater than a preset time T2; when the required consultation time T1 is greater than the preset time T2, sending a diagnosis request to all the doctor terminals in the doctor terminal set d4, and if no response information based on the diagnosis request is received within the preset time T2, sending a diagnosis request to the doctor terminal set d4; when the required consultation time T1 is not greater than the preset time T2, sending a diagnosis request to all the doctor terminals in the doctor terminal set d3, and if no response information based on the diagnosis request is received within the preset time T2, sending a diagnosis request to the doctor terminal set d4.

[0012] Further, the disease analysis report further includes an emergency parameter U; In the formula, T0 represents an initially set time threshold, U represents the emergency parameter, 0 < U < 1, and k1 represents an adjustable parameter used to control the influence degree of the emergency parameter on the time threshold.

[0013] Further, the doctor information further includes the total number N of online doctor terminals; the preset number N2 is:

[0014] N2 = k2N; in the formula, k2 represents an adjustable parameter used to control the influence degree of the total number N of online doctor terminals on the preset number N2.

[0015] In a second aspect, the present application provides a decision support system based on vital sign data, which adopts the decision support method based on vital sign data as in the first aspect, and includes:

[0016] A receiving module: which is used to receive patient data sent by a patient terminal, and the patient data includes vital sign data.

[0017] An analysis module: which is used to input patient data into a prediction model to obtain a disease analysis report of the patient; and obtain doctor information of a doctor terminal.

[0018] A matching module: which is used to match the disease analysis report of the patient with the doctor information to obtain a matched doctor terminal; and send a diagnosis request to the matched doctor terminal.

[0019] A decision support module: which is used to receive response information returned by a matched doctor based on the diagnosis request, and send the disease analysis report to the doctor terminal that returns the response information, so as to provide decision support to the doctor terminal.

[0020] In a third aspect, the present application provides a decision support device based on vital signs data, comprising: a memory and a processor; the memory is used to store a computer program; the processor is used to implement the steps of the decision support method based on vital signs data as in the first aspect when executing the computer program.

[0021] The technical solution provided by this application has at least the following technical effects or advantages:

[0022] By directly receiving vital signs data sent by the patient, and using predictive models to quickly generate disease analysis reports and send them to the doctor, decision support is provided to the doctor. This effectively solves the problem in existing technologies that it is difficult for doctors to quickly judge the patient's symptoms during online consultations. It reduces the time doctors spend manually collating and analyzing medical information, helps them better understand the patient's condition, and is conducive to doctors quickly judging the patient's symptoms and formulating more effective treatment plans, so that they can make decisions faster and improve diagnostic efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 A flowchart of a decision support method based on vital signs data in this application;

[0024] Figure 2 A schematic diagram of the process of matching the patient's disease analysis report with the doctor's information in this application. DETAILED DESCRIPTION

[0025] This application directly accepts vital signs data sent by the patient, uses a predictive model to quickly generate a disease analysis report and sends it to the doctor, providing decision support for the doctor, reducing the time doctors spend manually organizing and analyzing medical information, helping them better understand the patient's condition, and develop more effective treatment plans, so that they can make decisions faster. This efficiency improvement is particularly critical for the timely treatment of critically ill patients.

[0026] Embodiment 1:

[0027] like Figure 1 - Figure 2 As shown, this embodiment provides a decision support method based on vital sign data, which includes the following steps:

[0028] S100. Receive patient data sent by the patient end, where the patient data includes vital sign data.

[0029] First, you need to select a vital sign monitoring instrument that is compatible with the medical platform. Vital sign monitoring instruments include but are not limited to blood pressure monitors, blood glucose meters, electrocardiographs, etc. Vital sign monitors are usually equipped with Bluetooth or USB interfaces, which can be connected to smartphones or computers. Through the application or web interface provided by the platform, the device can be connected to the medical platform. Once the device is successfully connected to the medical platform, the measured vital sign data can be uploaded to the server of the medical platform through the device or application.

[0030] S200. Input patient data into the prediction model to obtain the patient's disease analysis report, which includes the disease type and the time required for consultation T1; obtain doctor information from the doctor side.

[0031] The prediction model has pattern recognition and prediction capabilities. It can establish the relationship between patient data and diseases by learning the patterns and features in a large amount of patient data, so as to make data-driven decisions and inferences and generate corresponding disease analysis reports. The model can mine the complex relationships and patterns in patient data, reveal potential disease diagnostic factors and risk indicators, and realize the automation and standardization of disease analysis, thereby improving the consistency and accuracy of diagnosis and prediction.

[0032] Patient data can also include the patient's medical history and symptom description. The medical history and symptom description can provide the patient's subjective feelings and changes in the development of the disease. The patient's medical history can reveal the cause, course and treatment history of the disease. Combining the medical history and symptom description can help to make personalized diagnosis and better understand the patient's disease characteristics and symptom manifestations.

[0033] In S200, inputting patient data into the prediction model specifically includes: inputting patient data into a multi-class logistic regression model to obtain disease types; and inputting patient data into a regression model to obtain the time required for consultation T1.

[0034] When training a multi-class logistic regression model, patient data is input, input data is cleaned, missing values ​​and outliers are processed, feature selection and feature engineering are performed, and data standardization or normalization and other preprocessing steps are performed. The formula of the logistic regression model can be defined as:

[0035] In the formula, P(Y=k|X) represents the predicted probability of suffering from different disease categories, X represents the input feature, β 0k ,β 1k ,…,β nk represents the intercept parameter of the kth category in the prediction model, β 0j ,β 1j ,…,β njRepresents the intercept parameter of the jth category in the prediction model, K represents the number of categories, k represents a specific category, and the model parameters are estimated using the training set. Usually, methods such as maximum likelihood estimation or gradient descent are used to evaluate the performance of the model on the test set, including accuracy, confusion matrix, classification report and other indicators for model evaluation. The trained multi-class logistic regression model is used to predict new patient data to obtain the most likely disease category.

[0036] After preprocessing the patient data, select features that affect the diagnosis time, such as patient characteristics, disease characteristics, etc., and use a linear regression model to build a model. Use the collected data to train the linear regression model, estimate the model's coefficients, and evaluate the model's goodness of fit to ensure the model's prediction accuracy for the diagnosis time. Use the trained linear regression model and the patient's characteristic data and substitute them into the model formula to predict the patient's diagnosis time.

[0037] Doctors can register themselves as doctor users of the platform through the registration process of the medical platform and provide relevant information such as personal information, qualification certificates and practice certificates. The platform can cooperate with major hospitals to obtain internal doctor information of the hospital, including doctor's title, department, professional field and other information, or cooperate with third-party data and professional association providers to obtain a doctor information database.

[0038] S300. Match the patient's disease analysis report with the doctor's information, obtain a matching doctor's terminal, and send a diagnosis request to the matching doctor's terminal. The doctor's information also includes the doctor's professional field, professional level and status information, and the status information includes the remaining time for consultation and idle status.

[0039] By matching the patient's disease analysis report with the doctor's professional field and level of expertise, we ensure that the diagnosis and treatment recommendations the patient receives come from professional doctors in related fields. Status information can help patients match appropriate doctors and ensure that doctors can diagnose and treat in a timely manner. This not only provides professional medical services, but also improves the patient's medical experience, allowing patients to obtain medical services more effectively.

[0040] Specifically, S300 includes the following steps: S310. Obtain doctor terminals that are in idle state and whose remaining consultation time is less than a preset time T2 and construct a doctor terminal set D1.

[0041] After obtaining the patient's disease analysis report, it is necessary to match the corresponding doctor based on the disease analysis report. The doctor end that is idle or has a remaining diagnosis time less than the preset time T2 can be matched. The doctor end whose remaining diagnosis time is less than the preset time T2 is the doctor who can complete the diagnosis faster. For the convenience of matching, these doctor ends can be put into the doctor end set D1.

[0042] S320. Match the disease type with the professional field of the doctors in the doctor-side set D1 to obtain the doctor-side set D2 that matches the disease type.

[0043] After clarifying the patient's disease type, traverse the doctor information in the doctor-side set D1, mainly their professional fields. For each doctor, check whether their professional field is related to the disease type. According to the doctor's professional field and disease type, use matching rules such as keyword matching and professional field code matching to filter out doctors that meet the matching rules to form the doctor-side set D2. Taking keyword matching as an example, it can be required that the number of disease keywords contained in the doctor's personal profile or professional field description reaches a certain threshold, or the match contains at least a certain keyword, and the doctor information that meets the matching rules is filtered out.

[0044] S330. Obtain the number N1 of doctor terminals in the doctor terminal set D2, and determine whether the number N1 is greater than the preset number N2. If so, execute matching mechanism one, otherwise execute matching mechanism two.

[0045] By obtaining the number N1 of doctor terminals in the doctor terminal set D2, the number N1 of doctor terminals that are currently or available in a short period of time can be known, and the current available medical resources can be understood in real time, specifically the number of doctors who are idle or can provide diagnosis and treatment services in a short period of time, so as to judge whether the current medical resources are tight, and the matching strategy can be selected according to the current medical resources. The dynamic matching strategy can be flexibly adjusted according to the actual medical resource situation, so as to better meet the patient's diagnosis and treatment needs and improve the utilization efficiency of medical resources.

[0046] Matching mechanism 1 includes: S331. Construct a doctor end set d1, put a preset number of doctors with the highest professional level in the doctor end set D2 into the doctor end set d1, and use the remaining doctors in the doctor end set D2 as the doctor end set d2; S332. Send a diagnosis request to all doctors in the doctor end set d1. If no response information based on the diagnosis request is received within the preset time T2, send a diagnosis request to the doctor end set d2.

[0047] If the number N1 is greater than the preset number N2, it means that the current supply of medical resources is sufficient. The matching mechanism 1 is adopted. By dividing all doctors who are currently or are available in the short term into two categories according to their professional level, the platform will give priority to matching doctors with high professional levels. Highly professional doctors usually have rich clinical experience and deep professional knowledge, and can provide more accurate diagnosis and more effective treatment plans, thereby maximizing the health and safety of patients. If no doctor responds to the diagnosis request within a period of time, the system will expand the matching range and send a diagnosis request to all doctors in the doctor set to ensure that patients can obtain diagnosis and treatment services in a timely manner. Through a flexible matching mechanism, the system can meet the needs of patients to the maximum extent according to actual conditions, effectively improving the utilization efficiency of medical resources.

[0048] In assessing the professionalism of doctors, the system can comprehensively consider multiple indicators such as the doctor's education, professional qualifications, clinical work experience, academic achievements, and patient evaluations. By recording the doctor's education, professional qualifications, years of work, number of clinical cases, patient cure rate, and other information, and considering the doctor's achievements in academic research and the patient's evaluation of the doctor, the system can establish a professionalism assessment file for each doctor. The weighted scores based on these indicators of the doctor can quantify the doctor's professionalism, so that doctors can be classified and matched according to the scores.

[0049] Matching mechanism 2 includes:

[0050] S331. Construct a doctor end set d3, put a preset number of doctor ends with the highest diagnosis efficiency DE in the doctor end set D2 into the doctor end set d3, and use the remaining doctor ends in the doctor end set D2 as the doctor end set d4.

[0051] S332. Determine whether the time required for consultation T1 is greater than the preset time T2.

[0052] S332a. When the time required for consultation T1 is greater than the preset time T2, a diagnosis request is sent to all doctor terminals in the doctor terminal set d4. If no response information based on the diagnosis request is received within the preset time T2, a diagnosis request is sent to the doctor terminal set d4.

[0053] S332b. When the time required for consultation T1 is not greater than the preset time T2, a diagnosis request is sent to all doctor terminals in the doctor terminal set d3. If no response information based on the diagnosis request is received within the preset time T2, a diagnosis request is sent to the doctor terminal set d4.

[0054] Classify the doctor terminals according to the diagnostic efficiency DE, and group the doctor terminals with high diagnostic efficiency into one category, which can better match the patient's medical needs and the doctor's diagnostic efficiency, thereby improving the utilization efficiency of medical resources. After judging whether the time T1 required for the patient to seek medical treatment is greater than the preset time T2, the platform will make a match according to different situations. If the time T1 required for the visit is greater than the preset time T2, it means that the patient needs a long time to seek medical treatment, and the doctor terminal with low diagnostic efficiency will be preferentially matched. If the time T1 required for the visit is less than the preset time T2, the doctor terminal with high diagnostic efficiency will be preferentially matched to maximize the utilization efficiency of medical resources and enable the doctor terminal with high diagnostic efficiency to match more patients.

[0055] The doctor terminals with high diagnostic efficiency are connected to patients who need short treatment times. In the same amount of time, the doctor terminals with high diagnostic efficiency can match more patients. In other words, these doctors with high diagnostic efficiency will have more opportunities to be assigned patients. If the quantity N1 is not greater than the preset quantity N2, it means that the supply of medical resources is tight at this time. In this case, doctors with high diagnostic efficiency can quickly complete the current diagnostic tasks, so that when patients in urgent need of medical treatment appear, they can free up time to handle emergencies more quickly, effectively reducing the waiting time of patients for treatment. It can better balance the distribution of medical resources, can maximize the satisfaction of patients' needs according to the actual situation, and improve the quality and efficiency of medical services.

[0056] The doctor information includes the diagnostic efficiency DE, and the diagnostic efficiency DE is: In the formula, DI i represents the evaluation effect DI of the doctor's i-th diagnosis, DT i represents the time spent by the doctor for the i-th diagnosis, and n represents the number of the doctor's diagnoses.

[0057] Combining the doctor's diagnostic effect and time provides a quantitative indicator to evaluate the doctor's diagnostic efficiency, enabling the doctor's performance to be objectively measured and compared. The medical platform can better understand the doctor's diagnostic efficiency, so as to more reasonably allocate patients to doctors with better diagnostic effects and shorter time spent, in order to improve the overall diagnostic efficiency, and can be used as an indicator for doctor performance evaluation, motivating doctors to strive to improve the diagnostic effect and at the same time try to save the time spent on diagnosis, thereby improving the quality and efficiency of the overall medical service.

[0058] The disease analysis report also includes an emergency parameter U; In the formula, T0 represents the initially set time threshold, U represents the emergency parameter, 0 < U < 1, and k1 represents an adjustable parameter used to control the influence degree of the emergency parameter on the time threshold.

[0059] The emergency parameter reflects the urgency of the patient's condition. The logistic regression model can also be used to predict the emergency parameter. The features related to the emergency parameter are selected as the input features of the model, and a continuous value between 0 and 1 is output as the value of the emergency parameter U. For example, vital signs include blood pressure, heart rate, respiratory rate and other indicators. The corresponding emergency parameter scores can be given according to the numerical range of these indicators. The data of these evaluation indicators are pre-processed, and then the model is trained to learn the relationship between the features and the emergency parameters in the sample data. After the training is completed, the probability of the patient's emergency parameter can be predicted, and the output emergency parameter value can be ensured to be between 0 and 1. The larger the value of U, the more urgent the patient's condition. When encountering an emergency patient, the system can automatically shorten the switching matching time while giving priority to matching the appropriate doctor, thereby ensuring that the patient can receive the corresponding medical services in a short time.

[0060] The doctor information also includes the total number N of online doctors; the preset number N2 is: N2=k2N; wherein k2 represents an adjustable parameter that controls the degree of influence of the total number N of online doctors on the preset number N2.

[0061] The number N2 represents the number of doctors who are idle or can quickly complete the current task, which is bound to the total number of online doctors. The total number N of online doctors can reflect the changes in medical service demand. When the medical service demand changes, there is no need to manually adjust N2, but automatically adjust the number of doctors who are idle or can quickly complete the current task according to the changes in the total number N of online doctors. When there are few doctors online, the platform will automatically reduce the number of doctors who are idle or can quickly complete the current task to avoid wasting resources. When there are many doctors online, the platform will also increase the number N2 accordingly to meet actual needs without human intervention, reducing management costs and complexity.

[0062] S400. Receive response information returned by a matched doctor based on the diagnosis request, and send a disease analysis report to the doctor end of the returned response information, thereby providing decision support to the doctor end.

[0063] By matching the right doctors, the system can ensure that patients receive targeted medical services, improve the personalization and professionalism of medical services, and the disease analysis report can provide the probability and diagnosis results of different diseases, provide doctors with more disease information and data support, and reduce the time doctors spend manually collating and analyzing disease information, helping them better understand the patient's condition and develop more effective treatment plans, so that they can make decisions faster.

[0064] Embodiment 2:

[0065] The present application provides a decision support system based on vital sign data, which adopts the decision support method based on vital sign data in the first embodiment, and includes: a receiving module, an analyzing module, a matching module and a decision support module.

[0066] The receiving module is used to receive patient data sent by the patient side, and the patient data includes vital signs data; the analysis module is used to input patient data into the prediction model to obtain the patient's disease analysis report; and obtain the doctor information of the doctor side; the matching module is used to match the patient's disease analysis report with the doctor information to obtain a matched doctor side, which is used to send a diagnosis request to the matched doctor side; the decision support module is used to receive the response information returned by the matched doctor based on the diagnosis request, and send the disease analysis report to the doctor side that returns the response information, thereby providing decision support to the doctor side.

[0067] This embodiment has all the advantages of the decision support method based on vital sign data, and can automatically implement all the steps of the decision support method based on vital sign data in the first embodiment.

[0068] In a third aspect, the present application provides a decision support device based on vital signs data, comprising: a memory and a processor; the memory is used to store a computer program; the processor is used to implement the steps of the decision support method based on vital signs data as in the first aspect when executing the computer program.

[0069] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented 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.

[0070] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps of the functions specified in a box or multiple boxes. Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they know the basic creative concept. Therefore, the attached claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

Claims

1. A decision support method based on vital signs data, characterized in that: It includes the following steps: Receiving patient data sent by a patient terminal, wherein the patient data includes vital sign data; Input patient data into the prediction model to obtain the patient's disease analysis report; obtain doctor information from the doctor's end; Matching the patient's disease analysis report with the doctor's information to obtain a matching doctor's terminal, and sending a diagnosis request to the matching doctor's terminal; Receive response information returned by a matched doctor based on the diagnosis request, and send the disease analysis report to the doctor end of the returned response information, thereby providing decision support to the doctor end.

2. The decision support method based on vital sign data according to claim 1, characterized in that: The disease analysis report includes the disease type and the time required for consultation T1; Input patient data into the prediction model to obtain the patient's disease analysis report, including: Input patient data into the multi-class logistic regression model to obtain disease type; Input patient data into the regression model to obtain the time required for consultation T1.

3. The decision support method based on vital sign data as claimed in claim 2, characterized in that: The doctor information includes the diagnosis efficiency DE, and the diagnosis efficiency DE is: Where, DI i Represents the evaluation effect of the doctor's i-th diagnosis DI, DT i represents the time taken by the doctor for the i-th diagnosis, and n represents the number of diagnoses made by the doctor.

4. The decision support method based on vital sign data as claimed in claim 3, characterized in that: The doctor information also includes the doctor's professional field, professional level and status information, and the status information includes the remaining consultation time and idle status; Matching the patient's disease analysis report with the doctor's information to obtain a matching doctor's end specifically includes: Obtain the doctor terminals that are in idle state and whose remaining consultation time is less than the preset time T2 and construct the doctor terminal set D1; Match the disease type with the professional field of doctors in the doctor-side set D1 to obtain a doctor-side set D2 that matches the disease type; Obtain the number N1 of doctor terminals in the doctor terminal set D2, and determine whether the number N1 of doctor terminals is greater than the preset number N2. If so, execute matching mechanism one, otherwise execute matching mechanism two.

5. The decision support method based on vital sign data as claimed in claim 4, characterized in that: The matching mechanism 1 includes: Construct a doctor end set d1, add a preset number of doctor ends with the highest professional degree in the doctor end set D2 into the doctor end set d1, and use the remaining doctor ends in the doctor end set D2 as the doctor end set d2; A diagnosis request is sent to all doctor terminals in the doctor terminal set d1. If no response information based on the diagnosis request is received within the preset time T2, a diagnosis request is sent to the doctor terminal set d2.

6. The decision support method based on vital sign data according to claim 5, characterized in that: The second matching mechanism includes: Construct a doctor end set d3, add a preset number of doctor ends with the highest diagnosis efficiency DE in the doctor end set D2 into the doctor end set d3, and use the remaining doctor ends in the doctor end set D2 as the doctor end set d4; Determine whether the time required for consultation T1 is greater than the preset time T2; When the time required for consultation T1 is greater than the preset time T2, a diagnosis request is sent to all doctor terminals in the doctor terminal set d4. If no response information based on the diagnosis request is received within the preset time T2, a diagnosis request is sent to the doctor terminal set d4; When the time required for a medical visit T1 is not greater than the preset time T2, a diagnosis request is sent to all doctor terminals in the doctor terminal set d3. If no response information based on the diagnosis request is received within the preset time T2, a diagnosis request is sent to the doctor terminal set d4.

7. The decision support method based on vital sign data according to claim 6, characterized in that: The disease analysis report further includes an emergency parameter U; Wherein, T0 represents the initially set time threshold, U represents the emergency parameter, 0 < U < 1, and k1 represents an adjustable parameter used to control the influence degree of the emergency parameter on the time threshold.

8. The decision support method based on vital sign data according to claim 4, characterized in that: The doctor information further includes the total number N of online doctor terminals; The preset number N2 is: N2 = k2N Wherein, k2 represents an adjustable parameter used to control the influence degree of the total number N of online doctor terminals on the preset number N2.

9. A decision support system based on vital signs data, which adopts the decision support method based on vital signs data as claimed in any one of claims 1 to 8, characterized in that: It includes: Receiving module: It is used to receive patient data sent by the patient terminal, and the patient data includes vital sign data; Analysis module: It is used to input patient data into the prediction model to obtain a disease analysis report of the patient; obtain doctor information of the doctor terminal; Matching module: It is used to match the disease analysis report of the patient with the doctor information to obtain a matching doctor terminal; Sending module: It is used to send a diagnosis request to the matching doctor terminal; Decision support module: It is used to receive the response information returned by the matching doctor based on the diagnosis request, and send the disease analysis report to the doctor terminal that returns the response information, so as to provide decision support to the doctor terminal.

10. A decision support device based on vital sign data, characterized in that: It includes: Memory, which is used to store computer programs; Processor, which is used to implement the steps of the decision support method based on vital sign data as described in any one of claims 1-8 when executing the computer program.

Images