WO2022012188A1 - 基于风险预测的就诊分配方法、装置、计算机设备 - Google Patents

基于风险预测的就诊分配方法、装置、计算机设备 Download PDF

Info

Publication number
WO2022012188A1
WO2022012188A1 PCT/CN2021/097209 CN2021097209W WO2022012188A1 WO 2022012188 A1 WO2022012188 A1 WO 2022012188A1 CN 2021097209 W CN2021097209 W CN 2021097209W WO 2022012188 A1 WO2022012188 A1 WO 2022012188A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
patient
risk
value
sample
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2021/097209
Other languages
English (en)
French (fr)
Inventor
唐蕊
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ping An Technology Shenzhen Co Ltd
Original Assignee
Ping An Technology Shenzhen Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Ping An Technology Shenzhen Co Ltd filed Critical Ping An Technology Shenzhen Co Ltd
Publication of WO2022012188A1 publication Critical patent/WO2022012188A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
    • G16H40/20—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Definitions

  • the present application relates to the technical field of intelligent decision-making, belongs to application scenarios related to smart medical care in smart cities, and in particular relates to a method, device, and computer equipment for medical treatment allocation based on risk prediction.
  • an individual's risk of illness can be assessed by introducing a health code including red, yellow, and green codes.
  • a health code including red, yellow, and green codes.
  • the inventor found that the health code generated by the above method can only make a rough prediction of the risk of disease, and since the generated health code is only based on the approximate recent activity trajectory of the individual, the basic information obtained is very limited. , it is impossible to conduct a comprehensive assessment of personal information; limited by the basic information obtained, the obtained disease risk prediction results are not refined enough, and individual differences cannot be accurately reflected in the assessment results, resulting in a weak assessment.
  • the embodiments of the present application provide a risk prediction-based medical visit allocation method, device, and computer equipment, which aim to solve the problem that medical visit allocation cannot be efficiently and accurately performed in the medical visit allocation method in the prior art method.
  • an embodiment of the present application provides a method for allocating medical visits based on risk prediction, which includes:
  • prediction request information from the client If the prediction request information from the client is received, obtain query information corresponding to the prediction type according to the pre-stored query database and the prediction type of the prediction request information, and send it to the client;
  • the reply information fed back by the client according to the inquiry information is received, the reply information and the prediction request information are quantized according to a pre-stored information quantification rule to obtain corresponding quantization information; wherein the quantization information includes Patient quantitative information and associated quantitative information;
  • medical visit allocation information corresponding to the risk prediction result and the prediction type is obtained according to the pre-stored current medical visit information, and fed back to the client.
  • an embodiment of the present application provides a risk prediction-based medical visit distribution device, which includes:
  • an inquiry information sending unit configured to obtain inquiry information corresponding to the prediction type according to a pre-stored inquiry database and the prediction type of the prediction request information and send it to the client if the prediction request information from the client is received ;
  • an information quantification unit configured to quantify the reply information and the prediction request information according to a pre-stored information quantification rule to obtain corresponding quantization information if the reply information fed back by the client according to the inquiry information is received;
  • the quantitative information includes patient quantitative information and associated quantitative information;
  • a patient risk value obtaining unit configured to obtain a patient risk value corresponding to the patient quantitative information according to a pre-stored patient risk prediction model
  • an associated risk value obtaining unit configured to obtain an associated risk value corresponding to the associated quantitative information according to a pre-stored associated risk prediction model
  • a risk prediction result obtaining unit used to calculate the patient risk value and the associated risk value according to a pre-stored risk coefficient calculation formula to obtain a corresponding risk prediction result and feed it back to the client;
  • the medical treatment allocation information obtaining unit is configured to obtain medical treatment allocation information corresponding to the risk prediction result and the prediction type according to the pre-stored current medical treatment information and feed back to the client if a medical treatment request is received from the client.
  • an embodiment of the present application further provides a computer device, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor executes the computer During the program, the method for assigning medical visits based on risk prediction described in the first aspect above is implemented.
  • an embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when executed by a processor, the computer program causes the processor to execute the above-mentioned first step.
  • the method for assigning medical visits based on risk prediction is described.
  • the embodiments of the present application provide a method, device, computer equipment and storage medium for medical visit allocation based on risk prediction.
  • Quantify the associated information to obtain the corresponding associated information obtain the patient risk value corresponding to the patient quantitative information according to the patient risk prediction model, and obtain the associated risk value corresponding to the associated risk according to the associated risk prediction model.
  • the risk prediction result according to the risk prediction result and prediction type, obtains the corresponding medical treatment assignment information and feeds it back to the client.
  • the corresponding patient risk value can be obtained according to the patient risk prediction model, and the corresponding associated risk value can be obtained according to the associated risk prediction model.
  • a more comprehensive assessment of the risk of a client's disease can be performed, and the accuracy of the obtained risk prediction result can be improved, thereby improving the efficiency and accuracy of patient allocation.
  • FIG. 1 is a schematic flowchart of a risk prediction-based medical visit allocation method provided by an embodiment of the present application
  • FIG. 2 is a schematic diagram of an application scenario of the risk prediction-based medical treatment allocation method provided by the embodiment of the present application;
  • FIG. 3 is a schematic diagram of a sub-flow of a method for assigning medical visits based on risk prediction provided by an embodiment of the present application;
  • FIG. 4 is a schematic diagram of another sub-flow of the method for assigning medical visits based on risk prediction provided by an embodiment of the present application;
  • FIG. 5 is another schematic flowchart of the method for assigning medical visits based on risk prediction according to an embodiment of the present application
  • FIG. 6 is a schematic diagram of another sub-flow of the method for assigning medical visits based on risk prediction provided by an embodiment of the present application;
  • FIG. 7 is another schematic flowchart of the method for assigning medical visits based on risk prediction provided by an embodiment of the present application.
  • FIG. 8 is a schematic diagram of another sub-flow of the method for assigning medical visits based on risk prediction provided by an embodiment of the present application.
  • FIG. 9 is a schematic block diagram of an apparatus for medical visit distribution based on risk prediction provided by an embodiment of the present application.
  • FIG. 10 is a schematic block diagram of a computer device provided by an embodiment of the present application.
  • FIG. 1 is a schematic flowchart of the risk prediction-based medical visit allocation method provided by an embodiment of the present application
  • FIG. 2 is a schematic diagram of an application scenario of the risk prediction-based medical visit allocation method provided by the embodiment of the present application.
  • the method for assigning medical visits based on risk prediction is applied in the management server 10 , and the method is executed by application software installed in the management server 10 .
  • the management server 10 communicates with at least one client 20 , and the client can input the input through the client 20 .
  • the prediction request information or reply information is sent to the management server 10 , and the management server 10 evaluates based on the received prediction request information or reply information to obtain a corresponding risk prediction result and feeds it back to the client 20 .
  • the management server 10 is an enterprise terminal used to execute the risk prediction-based medical visit allocation method to allocate medical visits to customers, and the client 20 is a terminal device that can be used to communicate with the management server 10, such as a desktop computer, a notebook computer, a tablet computer or mobile phone, etc. As shown in FIG. 1, the method includes steps S110-S160.
  • prediction request information from the client If the prediction request information from the client is received, obtain query information corresponding to the prediction type according to a pre-stored query database and the prediction type of the prediction request information, and send it to the client.
  • the query information corresponding to the prediction type is acquired according to the pre-stored query database and the prediction type of the prediction request information, and sent to the client.
  • the client can be a patient who needs to see a doctor after having certain symptoms.
  • the prediction request information is the request information entered by the client through the client to predict and score his/her own disease risk.
  • the prediction request information includes the prediction type and prediction.
  • the type is the type that needs to be predicted for a specific patient filled in by the user when inputting the prediction request information.
  • the prediction type may be a type of infectious diseases such as fever and respiratory infection.
  • the query database is a database pre-stored in the management server.
  • the query database includes multiple query data, the query database includes default query data, and also includes multiple types of query data corresponding to multiple prediction types. Since the information contained in the prediction request information is not comprehensive enough and the pertinence is not strong, the query information corresponding to the prediction type in the query database can be obtained and sent to the client.
  • step S110 includes sub-steps S111 and S112.
  • Each query data in the query database contains a corresponding type label.
  • the type label is the information recorded on the patient type or whether it is the default type corresponding to the query data. It can be obtained in the query database according to the label type of the query data. The first query data that matches the prediction type.
  • the first inquiry data corresponding to the tag information and "fever” can be obtained, and the first inquiry data can be the inquiry data for inquiring about the customer's current body temperature, whether to cough or not.
  • the default query data in the query database and the first query data are combined into the query information and sent to the client.
  • Obtain the query data whose type label is the default type in the query database as the default query data combine the default query data and the first query data into query information, and send the obtained query information to the client, and the combined query information can be It is reflected in the form of a questionnaire.
  • the default inquiry data may be inquiry data for inquiring about information such as the customer's travel use, past medical history, symptom description, symptom degree, and symptom duration.
  • the reply information fed back by the client according to the inquiry information is received, the reply information and the prediction request information are quantized according to a pre-stored information quantification rule to obtain corresponding quantization information, wherein the quantization information includes Patient quantitative information and associated quantitative information.
  • the prediction request information also includes the personal information of the customer, and the personal information may include information such as name, ID number, age, weight, height, address, office address (school address), and travel mode.
  • Information quantification rules are specific rules pre-stored in the management server for quantifying response information and prediction request information.
  • Information quantification rules include patient item information, associated item information, and item value quantification rules.
  • Patient item information is Information for recording items related to patients.
  • the items included in the patient item information can be the current body temperature, whether coughing, symptom description, symptom level and symptom duration; the associated item information is the information about other associated items.
  • the items included in the associated item information may be travel history, past medical history, age, weight, height, address, office address, and travel mode.
  • the information corresponding to the customer is classified according to the patient item information and related item information, and the classified patient information and related information are quantified according to the item value quantification rule, and the corresponding patient quantitative information and related quantitative information are obtained respectively.
  • step S120 includes sub-steps S121 and S122.
  • the associated information corresponding to the item information Specifically, the response information and the corresponding item value in the personal information are obtained according to the items included in the patient item information, the patient information corresponding to the patient item information can be obtained, and the individual is obtained according to the items included in the associated item information. Corresponding item values in the information can obtain the associated information corresponding to the associated item information.
  • the obtained related information of a certain customer is shown in Table 1.
  • S122 respectively perform the item value of the patient information and the item value of the associated information according to the item value quantification rule to obtain corresponding patient quantification information and associated quantification information.
  • the item value of the patient information and the item value of the associated information are respectively performed according to the item value quantification rule to obtain corresponding patient quantification information and associated quantification information.
  • the item value quantification rule includes a rule for converting each item value, each item value in the obtained patient quantification information and associated quantification information corresponds to a quantification value, and the quantification value corresponding to each item value is The range is [0, 1].
  • the corresponding conversion rule in the item value quantification rule is an activation function and an intermediate value
  • the corresponding quantification can be obtained by calculating the intermediate value and the item value of the item according to the activation function. value.
  • the activation function can be expressed as where x is the item value and v is the intermediate value corresponding to the item value.
  • the median value corresponding to age is 40, the age of a customer is 35, and the corresponding quantitative value is 0.2227 calculated according to the above activation function.
  • the corresponding conversion rule in the item value quantification rule is an information table containing multiple keywords and a value matching each keyword. Matching is performed, and the numerical value of the keyword matching the item value is obtained as the corresponding quantized value.
  • the conversion rule of travel mode contains multiple keywords "bus, subway, driving, walking", and the value corresponding to "subway” shown in Table 1 in the conversion rule is 0.4, then this value is regarded as " Metro" corresponds to the quantized value.
  • a patient risk value corresponding to the patient quantitative information is obtained according to a pre-stored patient risk prediction model.
  • the patient risk prediction model is a neural network that can predict the risk of patient quantitative information. Input the patient quantitative information into the patient risk prediction model to obtain a patient risk value, and the range of the patient risk value is [ 0, 1], the higher the patient risk value, the greater the risk of the client.
  • the patient risk prediction model includes multiple input nodes, two output nodes, and a fully connected hidden layer.
  • the input node is the node in the neural network that is used to input the quantitative information of a customer's patient.
  • the value is the input node value, then the output node value of each input node corresponds to a quantified value in the patient's quantitative information; the specific value of the output node is the output node value, and the two output node values correspond to the disease.
  • the probability that the risk is "yes” and the probability that the risk of disease is "no”, the softmax normalization operation of the two output node values is performed to obtain the corresponding two values, then the two values obtained after the normalization operation are the same.
  • the addition is 1; the normalized value of the probability value of the disease risk of "yes” is obtained as the corresponding patient risk value.
  • the fully connected hidden layer contains multiple feature units, and each feature unit is associated with all
  • the input node is associated with all output nodes.
  • Each feature unit corresponds to a feature unit value.
  • the feature unit value is the calculated value of the feature unit in the fully connected hidden layer.
  • the feature unit can be used to reflect the customer's patient quantitative information and the corresponding The association relationship between disease risks.
  • the association relationship can be reflected by the association formula between the feature unit and the input node or output node.
  • the association formula includes multiple parameters, and each parameter corresponds to a parameter value.
  • step S1310 is further included before step S130 .
  • the patient risk prediction model and the associated risk prediction model are respectively trained according to the sample data contained in the sample database and the pre-stored gradient calculation formula , to obtain the patient risk prediction model after training and the associated risk prediction model after training.
  • the administrator of the management server can input the sample database to train the patient risk prediction model and the associated risk prediction model, and the administrator can be a medical staff of the hospital.
  • the diagnostic information obtained by diagnosing the customer can be obtained at regular intervals. The diagnostic information is the information used to record whether the customer is diagnosed or not.
  • a sample database can be obtained, then each customer corresponds to a sample data in the sample database, and each sample data contains a sample Diagnostic information, as well as corresponding response information and personal information.
  • the gradient calculation formula is the calculation formula used when the model is trained by the gradient descent method.
  • step S1310 includes sub-steps S1311 , S1312 , S1313 , S1314 , S1315 and S1316 .
  • a piece of the sample data is quantified according to the information quantification rule to obtain corresponding sample quantification information, wherein the sample quantification information includes sample diagnosis quantification information, sample patient quantification information and sample correlation quantification information.
  • each piece of sample data contains reply information, personal information, and sample diagnostic information corresponding to the sample.
  • the process of quantifying the reply information and personal information in the sample data according to the information quantification rules is the same as the above-mentioned obtaining quantitative information.
  • the quantification of sample data also includes the quantification of sample diagnostic information.
  • the sample diagnostic quantification information can be obtained after quantifying the sample diagnostic information, and the corresponding sample quantification information can be obtained after quantifying the sample data including the sample diagnostic information.
  • a sample patient risk value corresponding to the sample patient quantitative information is obtained according to the patient risk prediction model.
  • the above process is the same as the process of obtaining the patient risk value, and will not be repeated here.
  • S1313 Acquire a sample-related risk value corresponding to the sample-related quantitative information according to the related risk prediction model.
  • a sample-related risk value corresponding to the sample-related quantitative information is obtained according to the associated risk prediction model.
  • the above process is the same as the process of obtaining the patient risk value, and will not be repeated here.
  • a first loss value corresponding to the sample patient risk value and a second loss value corresponding to the sample associated risk value are respectively obtained according to the sample diagnostic quantitative information.
  • the first loss value can be used to quantify the difference between the sample patient risk value and the sample diagnostic quantitative information
  • the second loss value can be used to quantify the difference between the sample associated risk value and the sample diagnostic quantitative information Express.
  • the first loss value or the second loss value can be obtained by calculating the loss function.
  • the loss function can be expressed as Among them, t is the sample patient risk value or the sample associated risk value, r is the sample diagnostic quantitative information, and f(t) is the calculated first loss value or second loss value.
  • the first loss value and the calculated value of the patient risk prediction model an update value of each parameter in the patient risk prediction model is calculated to update the parameter value of the parameter.
  • the calculated value obtained by calculating a parameter in the patient risk prediction model for the patient quantitative information of a certain sample is input into the gradient calculation formula, and combined with the above-mentioned first loss value, the corresponding parameter can be calculated. Update the value, this calculation process is also the gradient descent calculation.
  • the gradient calculation formula can be expressed as:
  • ⁇ x is the original parameter value of the parameter x
  • ⁇ is the preset learning rate in the gradient calculation formula
  • the parameter values of all parameters in the patient risk prediction model can be updated once, that is, the training of the patient risk prediction model is completed once;
  • the patient risk prediction model can be iteratively trained several times by using the sample data, so that the final patient risk prediction model is more accurate.
  • the second loss value and the calculated value of the associated risk prediction model the updated value of each parameter in the associated risk prediction model is calculated to update the parameter value of the parameter.
  • the specific process of updating the parameter value of each parameter in the associated risk prediction model is the same as the process of updating the parameter value of each parameter in the patient risk prediction model, and will not be repeated here.
  • the associated risk value corresponding to the associated quantitative information is acquired according to the pre-stored associated risk prediction model.
  • the associated risk prediction model is a neural network that can predict the risk of associated quantitative information. Entering the associated quantitative information into the associated risk prediction model can obtain an associated risk value.
  • the range of the associated risk value is [0, 1]. The higher the risk value, the greater the risk for the client.
  • the specific composition of the associated risk prediction model is the same as that of the patient risk prediction model, which will not be repeated here.
  • the patient risk value and the associated risk value are calculated according to a pre-stored risk coefficient calculation formula to obtain a corresponding risk prediction result, which is fed back to the client.
  • the risk coefficient calculation formula is the calculation formula pre-stored in the management server for calculating the customer's risk prediction result.
  • the risk prediction result can be represented by a score, and the calculated risk prediction result can be used for the customer from two aspects.
  • the comprehensive risk of the customer is predicted and scored. The higher the score in the risk prediction result, the greater the comprehensive risk of the client’s illness.
  • the range of the score in the obtained risk prediction result is [0, 100], and the obtained risk prediction result is fed back to the client terminal.
  • f 1 is the patient risk value and f 1 is the associated risk value.
  • step S150a is further included before step S150.
  • the weight information table is a data table used to store weight values in the management server.
  • the weight information table contains multiple groups of weight values, each group of weight values corresponds to a prediction type, and the corresponding prediction type can be obtained.
  • Configure the weight coefficient in the risk coefficient calculation formula to obtain the configured risk coefficient calculation formula.
  • medical visit allocation information corresponding to the risk prediction result and the prediction type is obtained according to the pre-stored current medical visit information, and fed back to the client.
  • the customer is a patient who needs to seek medical treatment, after viewing the risk prediction result, the customer can also send a medical consultation request to the associated server, and the management server can obtain the corresponding medical treatment allocation information according to the current medical treatment information, and feed it back to the client.
  • the medical appointment information can be used as a kind of appointment information. After receiving the medical appointment information, the customer can reach the corresponding destination according to the instructions of the medical appointment information and quickly seek medical treatment, thereby improving the customer's medical treatment efficiency.
  • the current medical treatment information is the information stored in the management server to record the medical treatment resources.
  • the current medical treatment information can be updated in real time.
  • the current medical treatment information includes multiple medical treatment areas, and each medical treatment area includes the corresponding medical treatment type and capacity. , if the number of patients in the clinic area does not exceed the clinic capacity, the clinic area is an idle clinic area, otherwise it is a non-idle clinic area.
  • step S160 includes sub-steps S161 and S162.
  • the priority allocation condition can be set as the prediction type is a strong infection type and the score in the risk prediction result is higher than 50 points.
  • the prediction type is a strong infection type
  • it can be determined whether the prediction type is the same as a certain type included in the strong infection type. If a keyword matches, it indicates that the prediction type belongs to the strong contagion type, otherwise it does not belong to the strong contagion type.
  • Each clinic area in the current medical information also includes the corresponding clinic address.
  • the clinic area that matches the address in the customer's personal information can be obtained as the pre-selected clinic area.
  • the clinic address can be obtained.
  • the clinic area with a distance of less than 2km from the address is used as the pre-selected clinic area; further judge the vacancy status of each clinic area in the pre-selected clinic area, and obtain all the free clinic areas in the pre-selected clinic area as the clinic assignment information and feed it back to the client.
  • Customers can select a suitable free clinic area from the clinic assignment information to make an appointment.
  • the customer can be included in the resource pool to be allocated, and the customer in the resource pool to be allocated can be allocated at a suitable time. For example, the resource to be allocated can be allocated the next day. If the customers in the pool are allocated, the customers in the resource pool to be allocated can be allocated to the specific time of the next day and in the clinic to make an appointment.
  • the technical methods in this application can be applied to application scenarios including smart government affairs/smart city management/smart community/smart security/smart logistics/smart medical care/smart education/smart environmental protection/smart transportation, etc. which include risk prediction-based medical treatment allocation, thereby promoting The construction of smart cities.
  • corresponding inquiry information is sent to the client according to the prediction request information from the client, and the feedback reply information is obtained, and the reply information and prediction are made according to the information quantification rules.
  • the patient information in the request information is quantified to obtain the corresponding patient quantitative information
  • the associated information is quantified to obtain the corresponding associated information
  • the patient risk value corresponding to the patient quantitative information is obtained according to the patient risk prediction model.
  • the associated risk prediction model obtains the associated risk value corresponding to the associated risk and then comprehensively obtains the risk prediction result, and obtains the corresponding medical treatment assignment information according to the risk prediction result and the prediction type and feeds it back to the client.
  • the corresponding patient risk value can be obtained according to the patient risk prediction model, and the corresponding associated risk value can be obtained according to the associated risk prediction model.
  • a more comprehensive assessment of the risk of a client's disease can be performed, and the accuracy of the obtained risk prediction result can be improved, thereby improving the efficiency and accuracy of patient allocation.
  • FIG. 9 is a schematic block diagram of an apparatus for medical visit distribution based on risk prediction provided by an embodiment of the present application.
  • the risk prediction-based medical visit distribution device may be configured in the management server 10 .
  • the risk prediction-based medical visit distribution device 100 includes an inquiry information sending unit 110, an information quantification unit 120, a patient risk value acquisition unit 130, a related risk value acquisition unit 140, a risk prediction result acquisition unit 150, and a medical visit distribution unit Information acquisition unit 160 .
  • the query information sending unit 110 is configured to obtain query information corresponding to the prediction type according to the pre-stored query database and the prediction type of the prediction request information and send it to the client if the prediction request information from the client is received end.
  • the inquiry information sending unit 110 includes subunits: a first inquiry data acquisition unit and an inquiry data combination unit.
  • a first query data acquisition unit configured to acquire the first query data matching the prediction type in the query database
  • a query data combination unit configured to combine the default query data in the query database with the first query data The query data is combined into the query information and sent to the client.
  • the information quantification unit 120 is configured to quantify the reply information and the prediction request information according to a pre-stored information quantification rule to obtain corresponding quantization information if the reply information fed back by the client according to the inquiry information is received.
  • the information quantization unit 120 includes subunits: an information classification unit and an item value quantization unit.
  • an information classification unit configured to classify the reply information and the personal information in the prediction request information according to the patient item information and the associated item information, so as to obtain a patient corresponding to the patient item information information and the associated information corresponding to the associated item information;
  • the item value quantification unit is used to separately perform the item value of the patient information and the item value of the associated information according to the item value quantification rule to obtain the corresponding Patient quantitative information and associated quantitative information.
  • the patient risk value obtaining unit 130 is configured to obtain a patient risk value corresponding to the patient quantitative information according to a pre-stored patient risk prediction model.
  • the risk prediction-based medical visit distribution apparatus 100 further includes a subunit: a model training unit.
  • the model training unit is used for, if the sample database input by the administrator of the management server is received, respectively, according to the sample data contained in the sample database and the pre-stored gradient calculation formula, the patient risk prediction model and the The associated risk prediction model is trained to obtain the trained patient risk prediction model and the trained associated risk prediction model.
  • the model training unit includes subunits: a sample quantitative information acquisition unit, a sample disease risk value acquisition unit, a sample associated risk value acquisition unit, a loss value acquisition unit, a first parameter value update unit, and a second parameter value update unit.
  • Parameter value update unit a sample quantitative information acquisition unit, a sample disease risk value acquisition unit, a sample associated risk value acquisition unit, a loss value acquisition unit, a first parameter value update unit, and a second parameter value update unit.
  • the sample quantification information acquisition unit is used to quantify a piece of the sample data according to the information quantification rule to obtain the corresponding sample quantification information; the sample disease risk value acquisition unit is used to obtain and obtain the data according to the patient risk prediction model.
  • a sample patient risk value corresponding to the sample patient quantification information a sample associated risk value obtaining unit, configured to obtain a sample associated risk value corresponding to the sample associated quantification information according to the associated risk prediction model;
  • a loss value obtaining unit for respectively obtaining the first loss value corresponding to the sample patient risk value and the second loss value corresponding to the sample associated risk value according to the sample diagnostic quantification information;
  • the first parameter value updating unit is used for According to the gradient calculation formula, the first loss value and the calculated value of the patient risk prediction model, the update value of each parameter in the patient risk prediction model is calculated to update the parameter value of the parameter;
  • a two-parameter value updating unit configured to calculate the updated value of each parameter in the associated risk prediction model according to the gradient calculation formula, the second loss value and the calculated value
  • the associated risk value obtaining unit 140 is configured to obtain the associated risk value corresponding to the associated quantitative information according to a pre-stored associated risk prediction model.
  • the risk prediction result obtaining unit 150 is configured to calculate the patient risk value and the associated risk value according to a pre-stored risk coefficient calculation formula to obtain a corresponding risk prediction result and feed it back to the client.
  • the risk prediction-based medical visit distribution apparatus 100 further includes a subunit: a weight coefficient configuration unit.
  • a weight coefficient configuration unit configured to obtain a set of weight values corresponding to the prediction type in a pre-stored weight information table according to the prediction type, and configure the weight coefficient in the risk coefficient calculation formula.
  • the medical treatment allocation information obtaining unit 160 is configured to obtain medical treatment allocation information corresponding to the risk prediction result and the prediction type according to the pre-stored current medical treatment information and feed it back to the client if a medical treatment request is received from the client .
  • the obtaining unit 160 for medical treatment assignment information includes subunits: a judging unit and a free clinic area obtaining unit.
  • a judging unit for judging whether the prediction type and the risk prediction result satisfy a preset priority allocation condition to obtain a judgment result
  • a free clinic area obtaining unit for obtaining the The free clinic area corresponding to the personal information of the prediction request information in the current medical visit information is fed back to the client as the medical visit allocation information.
  • the above-mentioned risk prediction-based medical visit distribution method is applied to the medical visit distribution device based on risk prediction provided by the embodiment of the present application, and the corresponding inquiry information is sent to the client according to the prediction request information from the client, and the feedback reply information is obtained.
  • the information quantification rules quantify the patient information in the reply information and the prediction request information to obtain the corresponding patient quantitative information, quantify the associated information to obtain the corresponding associated information, and obtain the patient quantitative information according to the patient risk prediction model.
  • the corresponding patient risk value, the associated risk value corresponding to the associated risk is obtained according to the associated risk prediction model, and then comprehensively obtains the risk prediction result. According to the risk prediction result and the prediction type, the corresponding medical appointment information is obtained and fed back to the client.
  • the corresponding patient risk value can be obtained according to the patient risk prediction model, and the corresponding associated risk value can be obtained according to the associated risk prediction model.
  • a more comprehensive assessment of the risk of a client's disease can be performed, and the accuracy of the obtained risk prediction result can be improved, thereby improving the efficiency and accuracy of patient allocation.
  • the above-mentioned apparatus for allocating medical visits based on risk prediction can be implemented in the form of a computer program, and the computer program can be executed on a computer device as shown in FIG. 10 .
  • FIG. 10 is a schematic block diagram of a computer device provided by an embodiment of the present application.
  • the computer equipment may be an enterprise terminal for executing a risk prediction-based medical visit allocation method to allocate medical visits to customers, for example, a server set in a hospital, a server set by a health service enterprise, and the like.
  • the computer device 500 includes a processor 502 , a memory and a network interface 505 connected by a system bus 501 , wherein the memory may include a non-volatile storage medium 503 and an internal memory 504 .
  • the nonvolatile storage medium 503 can store an operating system 5031 and a computer program 5032 .
  • the computer program 5032 when executed, can cause the processor 502 to perform a risk prediction based visit assignment method.
  • the processor 502 is used to provide computing and control capabilities to support the operation of the entire computer device 500 .
  • the internal memory 504 provides an environment for the execution of the computer program 5032 in the non-volatile storage medium 503.
  • the computer program 5032 can cause the processor 502 to execute a method for assigning medical visits based on risk prediction.
  • the network interface 505 is used for network communication, such as providing transmission of data information.
  • the network interface 505 is used for network communication, such as providing transmission of data information.
  • FIG. 10 is only a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device 500 to which the solution of the present application is applied.
  • the specific computer device 500 may include more or fewer components than shown, or combine certain components, or have a different arrangement of components.
  • the processor 502 is configured to run the computer program 5032 stored in the memory, so as to realize the corresponding functions in the above-mentioned risk prediction-based medical visit allocation method.
  • the embodiment of the computer device shown in FIG. 10 does not constitute a limitation on the specific structure of the computer device. Either some components are combined, or different component arrangements.
  • the computer device may only include a memory and a processor.
  • the structures and functions of the memory and the processor are the same as those of the embodiment shown in FIG. 10 , which will not be repeated here.
  • the processor 502 may be a central processing unit (Central Processing Unit, CPU), and the processor 502 may also be other general-purpose processors, digital signal processors (Digital Signal Processor, DSP), Application Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • the general-purpose processor can be a microprocessor or the processor can also be any conventional processor or the like.
  • a computer-readable storage medium may be a non-volatile computer-readable storage medium, or a volatile computer-readable storage medium.
  • the computer-readable storage medium stores a computer program, wherein when the computer program is executed by the processor, the steps included in the above-mentioned method for assigning medical visits based on risk prediction are implemented.
  • the disclosed apparatus, apparatus and method may be implemented in other manners.
  • the apparatus embodiments described above are only illustrative.
  • the division of the units is only logical function division.
  • there may be other division methods, or units with the same function may be grouped into one Units, such as multiple units or components, may be combined or may be integrated into another system, or some features may be omitted, or not implemented.
  • the shown or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may also be electrical, mechanical or other forms of connection.
  • the units described as separate components may or may not be physically separated, and components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solutions of the embodiments of the present application.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit.
  • the above-mentioned integrated units may be implemented in the form of hardware, or may be implemented in the form of software functional units.
  • the integrated unit if implemented in the form of a software functional unit and sold or used as an independent product, may be stored in a computer-readable storage medium.
  • a computer-readable storage medium includes several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application.
  • the aforementioned computer-readable storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), magnetic disk or optical disk and other media that can store program codes.

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Public Health (AREA)
  • Biomedical Technology (AREA)
  • Epidemiology (AREA)
  • Primary Health Care (AREA)
  • General Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Pathology (AREA)
  • Databases & Information Systems (AREA)
  • Business, Economics & Management (AREA)
  • General Business, Economics & Management (AREA)
  • Medical Treatment And Welfare Office Work (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

基于风险预测的就诊分配方法、装置、计算机设备及存储介质，该方法包括：根据来自客户端的预测请求信息发送对应的询问信息至该客户端并获取反馈的答复信息，根据信息量化规则对其中的病患信息进行量化得到对应的病患量化信息，对其中的关联信息进行量化得到对应的关联信息，根据病患量化信息获取对应的病患风险值、根据关联风险获取对应的关联风险值后进行综合得到风险预测结果，根据风险预测结果及预测类型获取对应的就诊分配信息并反馈至该客户端。该方法基于智能决策技术，属于人工智能技术领域，可通过两方面的信息对客户的患病风险进行评价，提高了所得到的风险预测结果的准确性，从而提高了对患者进行就诊分配的效率及准确率。

Description

基于风险预测的就诊分配方法、装置、计算机设备
本申请要求于2020年07月16日提交中国专利局、申请号为202010688314.0，发明名称为“基于风险预测的就诊分配方法、装置、计算机设备”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及智能决策技术领域，属于智慧城市智慧医疗相关的应用场景，尤其涉及一种基于风险预测的就诊分配方法、装置、计算机设备。
背景技术
流感流行时期为对每一个人的风险进行评估以对就诊患者进行快速区分并进行就诊分配，可通过引入包含红码、黄码、绿码的健康码对个人的患病风险进行评估。然而发明人发现，采用上述方式所生成的健康码仅能对患病风险进行一个大概的预测，且由于所生成的健康码仅基于个人近期大致的活动轨迹，因此所获取到的基本信息十分有限，无法对个人的信息进行全面评估；受限于所得到的基础信息，导致所得到的患病风险预测结果不够细化，无法在评估结果中精确体现个体性差异，导致评估的针对性不强、准确性不高，基于健康码所得到的评估结果无法准确、全面评估个人的患病风险从而影响了对患者进行就诊分配的效率及准确率。因此，现有的技术方法中的就诊分配方法存在无法高效、准确地进行就诊分配的问题。
发明内容
本申请实施例提供了一种基于风险预测的就诊分配方法、装置、计算机设备，旨在解决现有技术方法中的就诊分配方法所存在的无法高效、准确地进行就诊分配的问题。
第一方面，本申请实施例提供了一种基于风险预测的就诊分配方法，其包括：
若接收到来自所述客户端的预测请求信息，根据预存的询问数据库及所述预测请求信息的预测类型获取与所述预测类型对应的询问信息并发送至所述客户端；
若接收到所述客户端根据所述询问信息反馈的答复信息，根据预存的信息量化规则对所述答复信息及所述预测请求信息进行量化以得到对应的量化信息；其中，所述量化信息包括病患量化信息及关联量化信息；
根据预存的病患风险预测模型获取与所述病患量化信息对应的病患风险值；
根据预存的关联风险预测模型获取与所述关联量化信息对应的关联风险值；
根据预存的风险系数计算公式对所述病患风险值及所述关联风险值进行计算以得到对应的风险预测结果并反馈至所述客户端；
若接收到来自所述客户端的就诊请求，根据预存的当前就诊信息获取与所述风险预测结果及所述预测类型对应的就诊分配信息并反馈至所述客户端。
第二方面，本申请实施例提供了一种基于风险预测的就诊分配装置，其包括：
询问信息发送单元，用于若接收到来自所述客户端的预测请求信息，根据预存的询问数据库及所述预测请求信息的预测类型获取与所述预测类型对应的询问信息并发送至所述客户端；
信息量化单元，用于若接收到所述客户端根据所述询问信息反馈的答复信息，根据预存的信息量化规则对所述答复信息及所述预测请求信息进行量化以得到对应的量化信息；其中，所述量化信息包括病患量化信息及关联量化信息；
病患风险值获取单元，用于根据预存的病患风险预测模型获取与所述病患量化信息对应的病患风险值；
关联风险值获取单元，用于根据预存的关联风险预测模型获取与所述关联量化信息对应的关联风险值；
风险预测结果获取单元，用于根据预存的风险系数计算公式对所述病患风险值及所述关 联风险值进行计算以得到对应的风险预测结果并反馈至所述客户端；
就诊分配信息获取单元，用于若接收到来自所述客户端的就诊请求，根据预存的当前就诊信息获取与所述风险预测结果及所述预测类型对应的就诊分配信息并反馈至所述客户端。
第三方面，本申请实施例又提供了一种计算机设备，其包括存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现上述第一方面所述的基于风险预测的就诊分配方法。
第四方面，本申请实施例还提供了一种计算机可读存储介质，其中所述计算机可读存储介质存储有计算机程序，所述计算机程序当被处理器执行时使所述处理器执行上述第一方面所述的基于风险预测的就诊分配方法。
本申请实施例提供了一种基于风险预测的就诊分配方法、装置、计算机设备及存储介质。根据来自客户端的预测请求信息发送对应的询问信息至该客户端，并获取反馈的答复信息，根据信息量化规则对答复信息及预测请求信息中的病患信息进行量化得到对应的病患量化信息，对其中的关联信息进行量化得到对应的关联信息，根据病患风险预测模型获取与病患量化信息对应的病患风险值、根据关联风险预测模型获取与关联风险对应的关联风险值后进行综合得到风险预测结果，根据风险预测结果及预测类型获取对应的就诊分配信息并反馈至该客户端。通过上述方法，可依据病患风险预测模型获取对应的病患风险值、根据关联风险预测模型获取对应的关联风险值，通过两方面的信息对客户的患病风险更有针对性地进行评价，可更加全面地对客户的患病风险进行评估，并提高所得到的风险预测结果的准确性，从而提高了对患者进行就诊分配的效率及准确率。
附图说明
为了更清楚地说明本申请实施例技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1为本申请实施例提供的基于风险预测的就诊分配方法的流程示意图；
图2为本申请实施例提供的基于风险预测的就诊分配方法的应用场景示意图；
图3为本申请实施例提供的基于风险预测的就诊分配方法的子流程示意图；
图4为本申请实施例提供的基于风险预测的就诊分配方法的另一子流程示意图；
图5为本申请实施例提供的基于风险预测的就诊分配方法的另一流程示意图；
图6为本申请实施例提供的基于风险预测的就诊分配方法的另一子流程示意图；
图7为本申请实施例提供的基于风险预测的就诊分配方法的另一流程示意图；
图8为本申请实施例提供的基于风险预测的就诊分配方法的另一子流程示意图；
图9为本申请实施例提供的基于风险预测的就诊分配装置的示意性框图；
图10为本申请实施例提供的计算机设备的示意性框图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
应当理解，当在本说明书和所附权利要求书中使用时，术语“包括”和“包含”指示所描述特征、整体、步骤、操作、元素和/或组件的存在，但并不排除一个或多个其它特征、整体、步骤、操作、元素、组件和/或其集合的存在或添加。
还应当理解，在此本申请说明书中所使用的术语仅仅是出于描述特定实施例的目的而并不意在限制本申请。如在本申请说明书和所附权利要求书中所使用的那样，除非上下文清楚地指明其它情况，否则单数形式的“一”、“一个”及“该”意在包括复数形式。
还应当进一步理解，在本申请说明书和所附权利要求书中使用的术语“和/或”是指相关联列出的项中的一个或多个的任何组合以及所有可能组合，并且包括这些组合。
请参阅图1及图2，图1是本申请实施例提供的基于风险预测的就诊分配方法的流程示意图，图2为本申请实施例提供的基于风险预测的就诊分配方法的应用场景示意图。该基于风险预测的就诊分配方法应用于管理服务器10中，该方法通过安装于管理服务器10中的应用软件进行执行，管理服务器10与至少一台客户端20进行通信，客户可通过客户端20输入预测请求信息或答复信息发送至管理服务器10，管理服务器10基于所接收到的预测请求信息或答复信息进行评价得到对应的风险预测结果并反馈至客户端20。管理服务器10即是用于执行基于风险预测的就诊分配方法以对客户进行就诊分配的企业终端，客户端20即是可用于与管理服务器10进行通信的终端设备，例如台式电脑、笔记本电脑、平板电脑或手机等。如图1所示，该方法包括步骤S110～S160。
S110、若接收到来自所述客户端的预测请求信息，根据预存的询问数据库及所述预测请求信息的预测类型获取与所述预测类型对应的询问信息并发送至所述客户端。
若接收到来自所述客户端的预测请求信息，根据预存的询问数据库及所述预测请求信息的预测类型获取与所述预测类型对应的询问信息并发送至所述客户端。客户可以是出现某些症状后需要进行就诊的患者，预测请求信息即为客户通过客户端所输入的用于对其自身患病风险进行预测评分的请求信息，预测请求信息中包含预测类型，预测类型即为用户输入预测请求信息时所填写的需对具体病患进行预测的类型，例如，预测类型可以是发热、呼吸道感染等传染病类型。询问数据库即为管理服务器中预存的一数据库，询问数据库中包含多条询问数据，询问数据库中包括默认询问数据，还包括与多个预测类型对应的多类询问数据。由于预测请求信息中所包含的信息不够全面、针对性不强，可获取询问数据库中与预测类型对应的询问信息并发送至客户端。
在一实施例中，如图3所示，步骤S110包括子步骤S111和S112。
S111、获取所述询问数据库中与所述预测类型相匹配的第一询问数据。
获取所述询问数据库中与所述预测类型相匹配的第一询问数据。询问数据库中的每一条询问数据均对应包含一个类型标签，类型标签即是对一条询问数据所对应的病患类型或是否为默认类型进行记录的信息，可根据询问数据的标签类型获取询问数据库中与预测类型相匹配的第一询问数据。
例如，预测类型为“发热”，则可获取标签信息与“发热”对应的第一询问数据，则第一询问数据可以是对客户的当前体温、是否咳嗽等信息进行询问的提问数据。
S112、将所述询问数据库中的默认询问数据及所述第一询问数据组合为所述询问信息并发送至所述客户端。
将所述询问数据库中的默认询问数据及所述第一询问数据组合为所述询问信息并发送至所述客户端。获取询问数据库中类型标签为默认类型的询问数据作为默认询问数据，并将默认询问数据与第一询问数据组合为询问信息，将所得到的询问信息发送至客户端，所组合得到的询问信息可以是以问卷方式进行体现。默认询问数据可以是对客户的旅行使、既往病史、症状描述、症状程度及症状持续时间等信息进行询问的提问数据。
S120、若接收到所述客户端根据所述询问信息反馈的答复信息，根据预存的信息量化规则对所述答复信息及所述预测请求信息进行量化以得到对应的量化信息。
若接收到所述客户端根据所述询问信息反馈的答复信息，根据预存的信息量化规则对所述答复信息及所述预测请求信息进行量化以得到对应的量化信息，其中，所述量化信息包括病患量化信息及关联量化信息。预测请求信息中还包含客户的个人信息，个人信息可以包括姓名、身份证号、年龄、体重、身高、住址、办公地址(上学地址)、出行方式等信息。信息量化规则即为管理服务器中预先存储的用于对答复信息及预测请求信息进行量化的具体规则， 信息量化规则包括病患项目信息、关联项目信息及项目值量化规则，病患项目信息即是对与病患相关的项目进行记录的信息，例如，病患项目信息中所包含的项目可以是当前体温、是否咳嗽、症状描述、症状程度及症状持续时间；关联项目信息即为对其他相关联的项目进行记录的信息，例如，关联项目信息中所包含的项目可以是旅行史、既往病史、年龄、体重、身高、住址、办公地址及出行方式。根据病患项目信息及关联项目信息对客户对应的信息进行分类，并根据项目值量化规则对分类得到的病患信息及关联信息进行量化，分别得到对应的病患量化信息及关联量化信息。
在一实施例中，如图4所示，步骤S120包括子步骤S121和S122。
S121、根据所述病患项目信息及所述关联项目信息对所述答复信息及所述预测请求信息中的个人信息进行分类，以得到与所述病患项目信息对应的病患信息及与所述关联项目信息对应的关联信息。
根据所述病患项目信息及所述关联项目信息对所述答复信息及所述预测请求信息中的个人信息进行分类，以得到与所述病患项目信息对应的病患信息及与所述关联项目信息对应的关联信息。具体的，根据病患项目信息中所包含的项目获取答复信息及个人信息中对应的项目值，即可得到与病患项目信息对应的病患信息，根据关联项目信息中所包含的项目获取个人信息中对应的项目值，即可得到与关联项目信息对应的关联信息。
例如，所得到的某一客户的关联信息如表1所示。
Figure PCTCN2021097209-appb-000001
表1
S122、根据所述项目值量化规则对所述病患信息的项目值及所述关联信息的项目值分别进行以得到对应的病患量化信息及关联量化信息。
根据所述项目值量化规则对所述病患信息的项目值及所述关联信息的项目值分别进行以得到对应的病患量化信息及关联量化信息。为对所得到的病患信息及关联信息中做包含的每一项目值进行量化，需通过项目值量化规则将相应项目值转换为对应的量化值，量化值即可对与每一项目对应的项目值进行量化表示。具体的，项目值量化规则中包括对每一项目值进行转换的规则，所得到的病患量化信息及关联量化信息中每一项目值均对应一个量化值，每一项目值对应的量化值的范围均为[0，1]。
具体的，对于以数值方式表示的项目值，项目值量化规则中对应的转换规则为一个激活函数及一个中间值，根据激活函数对中间值及项目的项目值进行计算，即可得到对应的量化值。
例如，激活函数可表示为
Figure PCTCN2021097209-appb-000002
其中，x为项目值，v为与该项目值对应的中间值。与年龄对应的中间值为40，某客户的年龄为35，根据上述激活函数计算得到对应的量化值为0.2227。
对于以非数值方式进行表示的项目值，项目值量化规则中对应的转换规则为一个包含多个关键字以及与每一关键字相匹配的数值的信息表，根据项目值中与每一关键字进行匹配，并获取与项目值相匹配的关键字的数值作为对应的量化值。
例如，出行方式的转换规则中包含多个关键字“公交、地铁、驾车、步行”，该转换规则中与表1中所示的“地铁”对应的数值为0.4，则将该数值作为与“地铁”对应的量化值。
S130、根据预存的病患风险预测模型获取与所述病患量化信息对应的病患风险值。
根据预存的病患风险预测模型获取与所述病患量化信息对应的病患风险值。病患风险预测模型即为一个可对病患量化信息进行风险预测的神经网络，将病患量化信息输入该病患风 险预测模型即可得到一个病患风险值，病患风险值的范围为[0，1]，病患风险值越高则表明该客户患病风险越大。其中，病患风险预测模型包括多个输入节点、两个输出节点及全连接隐层，输入节点即是神经网络中用于对某一客户的病患量化信息进行输入的节点，输入节点的具体数值即为输入节点值，则每一输入节点的输出节点值即与病患量化信息中的一个量化值相对应；输出节点的具体数值即为输出节点值，两个输出节点值分别对应患病风险为“是”的概率及患病风险为“否”的概率，将两个输出节点值进行softmax归一化操作得到对应的两个数值，则归一化操作后所得到的两个数值相加即为1；获取患病风险为“是”的概率值进行归一化操作后的数值作为对应的病患风险值，全连接隐层中包含多个特征单元，每一个特征单元均与所有输入节点和所有输出节点进行关联，每一特征单元均对应一个特征单元值，特征单元值即为全连接隐层中的特征单元的计算值，特征单元可用于反映客户的病患量化信息与对应患病风险之间的关联关系，关联关系可通过特征单元与输入节点或输出节点之间的关联公式进行体现，关联公式中包含多个参数，每一参数对应一个参数值。
在一实施例中，如图5所示，步骤S130之前还包括步骤S1310。
S1310、若接收到所述管理服务器的管理员所输入的样本数据库，根据所述样本数据库所包含的样本数据及预存的梯度计算公式分别对所述病患风险预测模型及所述关联风险预测模型进行训练，以得到训练后的所述病患风险预测模型及训练后的所述关联风险预测模型。
若接收到所述管理服务器的管理员所输入的样本数据库，根据所述样本数据库所包含的样本数据及预存的梯度计算公式分别对所述病患风险预测模型及所述关联风险预测模型进行训练，以得到训练后的所述病患风险预测模型及训练后的所述关联风险预测模型。管理服务器的管理员可输入样本数据库对病患风险预测模型及关联风险预测模型进行训练，管理员可以是医院的医护人员。在应用使用过程中，对于SARS或新冠肺炎等传染性疾病，每隔一段时间均可获取对客户进行诊断所得到的诊断信息，诊断信息即是用于对客户是否确诊进行记录的信息，获取到一批客户的诊断信息后，与该批次客户的答复信息及个人信息进行组合即可得到一个样本数据库，则每一个客户在样本数据库中均对应一条样本数据，每一条样本数据均包含一个样本诊断信息，以及对应的答复信息及个人信息。其中，梯度计算公式即为通过梯度下降方法对模型进行训练时所采用的计算公式。
在一实施例中，如图6所示，步骤S1310包括子步骤S1311、S1312、S1313、S1314、S1315和S1316。
S1311、根据所述信息量化规则对一条所述样本数据进行量化以得到对应的样本量化信息。
根据所述信息量化规则对一条所述样本数据进行量化以得到对应的样本量化信息，其中，所述样本量化信息包括样本诊断量化信息、样本病患量化信息及样本关联量化信息。具体的，每一条样本数据中均包含与该样本对应的答复信息、个人信息及样本诊断信息，则根据信息量化规则对样本数据中的答复信息及个人信息进行量化的过程即与上述获取量化信息的过程相同，除此之外，对样本数据进行量化过程中还包含对样本诊断信息的量化，若样本诊断信息为“是”，则对应的量化值为“1”；若样本诊断信息为“否”，则对应的量化值为“0”，对样本诊断信息进行量化后即可得到样本诊断量化信息，对包含样本诊断信息的样本数据进行量化后即可得到对应的样本量化信息。
S1312、根据所述病患风险预测模型获取与所述样本病患量化信息对应的样本病患风险值。
根据所述病患风险预测模型获取与所述样本病患量化信息对应的样本病患风险值。上述过程与获取病患风险值的过程相同，在此不作赘述。
S1313、根据所述关联风险预测模型获取与所述样本关联量化信息对应的样本关联风险值。
根据所述关联风险预测模型获取与所述样本关联量化信息对应的样本关联风险值。上述过程与获取病患风险值的过程相同，在此不作赘述。
S1314、根据所述样本诊断量化信息分别获取与所述样本病患风险值对应的第一损失值及 与所述样本关联风险值对应的第二损失值。
根据所述样本诊断量化信息分别获取与所述样本病患风险值对应的第一损失值及与所述样本关联风险值对应的第二损失值。具体的，第一损失值可用于对样本病患风险值与样本诊断量化信息之间的差异进行量化表示，第二损失值可用于对样本关联风险值与样本诊断量化信息之间的差异进行量化表示。具体的，可通过损失函数计算得到第一损失值或第二损失值。
例如，损失函数可表示为
Figure PCTCN2021097209-appb-000003
其中，t为样本病患风险值或样本关联风险值，r为样本诊断量化信息，f(t)即为所计算得到的第一损失值或第二损失值。
S1315、根据所述梯度计算公式、所述第一损失值及所述病患风险预测模型的计算值计算得到所述病患风险预测模型中每一参数的更新值以更新所述参数的参数值。
根据所述梯度计算公式、所述第一损失值及所述病患风险预测模型的计算值计算得到所述病患风险预测模型中每一参数的更新值以更新所述参数的参数值。具体的，将病患风险预测模型中一个参数对某一样本的病患量化信息进行计算所得到的计算值输入梯度计算公式，并结合上述第一损失值，即可计算得到与该参数对应的更新值，这一计算过程也即为梯度下降计算。
具体的，梯度计算公式可表示为：
Figure PCTCN2021097209-appb-000004
其中，
Figure PCTCN2021097209-appb-000005
为计算得到的参数x的更新值，ω x为参数x的原始参数值，η为梯度计算公式中预置的学习率，
Figure PCTCN2021097209-appb-000006
为基于第一损失值及参数x对应的计算值对该参数x的偏导值(这一计算过程中需使用参数对应的计算值)。
根据一条样本数据对应的第一损失值即可对病患风险预测模型中所有参数的参数值进行一次更新，也即是完成对病患风险预测模型的一次训练；根据样本数据库中所包含的多条样本数据即可对病患风险预测模型进行多次迭代训练，以使最终得到的病患风险预测模型更加准确。
S1316、根据所述梯度计算公式、所述第二损失值及所述关联风险预测模型的计算值计算得到所述关联风险预测模型中每一参数的更新值以更新所述参数的参数值。
根据所述梯度计算公式、所述第二损失值及所述关联风险预测模型的计算值计算得到所述关联风险预测模型中每一参数的更新值以更新所述参数的参数值。对关联风险预测模型中每一参数的参数值进行更新的具体过程，与对病患风险预测模型中每一参数的参数值进行更新的过程相同，在此不作赘述。
S140、根据预存的关联风险预测模型获取与所述关联量化信息对应的关联风险值。
根据预存的关联风险预测模型获取与所述关联量化信息对应的关联风险值。关联风险预测模型即为一个可对关联量化信息进行风险预测的神经网络，将关联量化信息输入该关联风险预测模型即可得到一个关联风险值，关联风险值的范围为[0，1]，关联风险值越高则表明该客户患病风险越大。关联风险预测模型的具体构成与病患风险预测模型相同，在此不作赘述。
S150、根据预存的风险系数计算公式对所述病患风险值及所述关联风险值进行计算以得到对应的风险预测结果并反馈至所述客户端。
根据预存的风险系数计算公式对所述病患风险值及所述关联风险值进行计算以得到对应的风险预测结果并反馈至所述客户端。风险系数计算公式即为管理服务器中预存的用于计算客户的风险预测结果的计算公式，风险预测结果可采用一个评分进行表示，则所计算得到的风险预测结果即可从两个方面对该客户的综合风险进行预测评分，风险预测结果中评分越高则表明该客户患病的综合风险越大，所得到的风险预测结果中评分的范围为[0,100]，将所得到的风险预测结果反馈至该客户终端。其中，风险系数计算公式可以是z＝100×(p×f 1+q×f 2)，其中，p及q均为该风险系数计算公式中预设的权重系数，p+q＝1，z为计算得到的评分，f 1 为病患风险值，f 1为关联风险值。
在一实施例中，如图7所示，步骤S150之前还包括步骤S150a。
S150a、根据所述预测类型获取预存的权重信息表中与所述预测类型对应的一组权重值，对所述风险系数计算公式中的权重系数进行配置。具体的，权重信息表即为管理服务器中用于存储权重值的数据表，权重信息表中包含多组权重值，每一组权重值均与一个预测类型相对应，则可根据预测类型获取对应的一组权重值对风险系数计算公式中的权重系数进行配置，以得到配置后的风险系数计算公式，通过对风险系数计算公式中的权重系数进行配置，可进一步增加对风险预测结果进行计算的准确性。
S160、若接收到来自所述客户端的就诊请求，根据预存的当前就诊信息获取与所述风险预测结果及所述预测类型对应的就诊分配信息并反馈至所述客户端。
若接收到来自所述客户端的就诊请求，根据预存的当前就诊信息获取与所述风险预测结果及所述预测类型对应的就诊分配信息并反馈至所述客户端。若该客户为需要进行就诊的患者，则该客户查看到风险预测结果后，还可向关联服务器发送就诊请求，管理服务器可根据当前就诊信息获取对应的就诊分配信息，并反馈至该客户端，则就诊分配信息可作为一种预约信息，客户获接收到就诊分配信息后即可根据该就诊分配信息的指示到达相应目的地并进行快速就诊，提高客户的就诊效率。当前就诊信息即为管理服务器中所存储的对就诊资源进行记录的信息，当前就诊信息可实时更新，当前就诊信息中包含多个诊区，每一诊区中均包含对应的就诊类型及就诊容量，若诊区的就诊人数未超过就诊容量，则该诊区为空闲诊区，否则为非空闲诊区。
在一实施例中，如图8所示，步骤S160包括子步骤S161和S162。
S161、对所述预测类型及所述风险预测结果是否满足预置的优先分配条件进行判断以得到判断结果。
首先可判断预测类型及风险预测结果是否满足优先分配条件，若满足则进行优先分配。例如，可设置优先分配条件为预测类型为强传染类型且风险预测结果中评分高于50分，在判断预测类型是否为强传染类型时，可判断预测类型是否与强传染类型中所包含的某一关键字相匹配，若相匹配则表明该预测类型属于强传染类型，否则不属于强传染类型。
S162、若所述判断结果为是，获取所述当前就诊信息中与所述预测请求信息的个人信息对应的空闲诊区作为所述就诊分配信息反馈至所述客户端。
若所述判断结果为是，则对该客户进行优先分配。当前就诊信息中每一诊区还包含对应的诊区地址，可根据每一诊区的就诊地址获取与该客户的个人信息中住址相匹配的诊区作为预选诊区，例如，可获取就诊地址与住址之间距离在2km以内的诊区作为预选诊区；进一步判断预选诊区中每一诊区的空闲状态，并获取预选诊区中所有空闲诊区作为就诊分配信息反馈至该客户端，客户可从就诊分配信息中选择一个合适的空闲诊区进行预约就诊。若判断结果为否，则不对该客户进行优先分配，可将该客户收入待分配资源池中，在合适时间点在对待分配资源池中的客户进行分配，例如，可在第二天对待分配资源池中的客户进行分配，则可将待分配资源池中的客户分配至第二天的具体时间及诊区内进行预约就诊。
本申请中的技术方法可应用于智慧政务/智慧城管/智慧社区/智慧安防/智慧物流/智慧医疗/智慧教育/智慧环保/智慧交通等包含基于风险预测的就诊分配的应用场景中，从而推动智慧城市的建设。
在本申请实施例所提供的基于风险预测的就诊分配方法中，根据来自客户端的预测请求信息发送对应的询问信息至该客户端，并获取反馈的答复信息，根据信息量化规则对答复信息及预测请求信息中的病患信息进行量化得到对应的病患量化信息，对其中的关联信息进行量化得到对应的关联信息，根据病患风险预测模型获取与病患量化信息对应的病患风险值、根据关联风险预测模型获取与关联风险对应的关联风险值后进行综合得到风险预测结果，根 据风险预测结果及预测类型获取对应的就诊分配信息并反馈至该客户端。通过上述方法，可依据病患风险预测模型获取对应的病患风险值、根据关联风险预测模型获取对应的关联风险值，通过两方面的信息对客户的患病风险更有针对性地进行评价，可更加全面地对客户的患病风险进行评估，并提高所得到的风险预测结果的准确性，从而提高了对患者进行就诊分配的效率及准确率。
本申请实施例还提供一种基于风险预测的就诊分配装置，该基于风险预测的就诊分配装置用于执行前述基于风险预测的就诊分配方法的任一实施例。具体地，请参阅图9，图9是本申请实施例提供的基于风险预测的就诊分配装置的示意性框图。该基于风险预测的就诊分配装置可以配置于管理服务器10中。
如图9所示，基于风险预测的就诊分配装置100包括询问信息发送单元110、信息量化单元120、病患风险值获取单元130、关联风险值获取单元140、风险预测结果获取单元150和就诊分配信息获取单元160。
询问信息发送单元110，用于若接收到来自所述客户端的预测请求信息，根据预存的询问数据库及所述预测请求信息的预测类型获取与所述预测类型对应的询问信息并发送至所述客户端。
在一实施例中，所述询问信息发送单元110包括子单元：第一询问数据获取单元及询问数据组合单元。
第一询问数据获取单元，用于获取所述询问数据库中与所述预测类型相匹配的第一询问数据；询问数据组合单元，用于将所述询问数据库中的默认询问数据及所述第一询问数据组合为所述询问信息并发送至所述客户端。
信息量化单元120，用于若接收到所述客户端根据所述询问信息反馈的答复信息，根据预存的信息量化规则对所述答复信息及所述预测请求信息进行量化以得到对应的量化信息。
在一实施例中，所述信息量化单元120包括子单元：信息分类单元及项目值量化单元。
信息分类单元，用于根据所述病患项目信息及所述关联项目信息对所述答复信息及所述预测请求信息中的个人信息进行分类，以得到与所述病患项目信息对应的病患信息及与所述关联项目信息对应的关联信息；项目值量化单元，用于根据所述项目值量化规则对所述病患信息的项目值及所述关联信息的项目值分别进行以得到对应的病患量化信息及关联量化信息。
病患风险值获取单元130，用于根据预存的病患风险预测模型获取与所述病患量化信息对应的病患风险值。
在一实施例中，所述基于风险预测的就诊分配装置100还包括子单元：模型训练单元。
模型训练单元，用于若接收到所述管理服务器的管理员所输入的样本数据库，根据所述样本数据库所包含的样本数据及预存的梯度计算公式分别对所述病患风险预测模型及所述关联风险预测模型进行训练，以得到训练后的所述病患风险预测模型及训练后的所述关联风险预测模型。
在一实施例中，所述模型训练单元包括子单元：样本量化信息获取单元、样本患病风险值获取单元、样本关联风险值获取单元、损失值获取单元、第一参数值更新单元及第二参数值更新单元。
样本量化信息获取单元，用于根据所述信息量化规则对一条所述样本数据进行量化以得到对应的样本量化信息；样本患病风险值获取单元，用于根据所述病患风险预测模型获取与所述样本病患量化信息对应的样本病患风险值；样本关联风险值获取单元，用于根据所述关联风险预测模型获取与所述样本关联量化信息对应的样本关联风险值；损失值获取单元，用于根据所述样本诊断量化信息分别获取与所述样本病患风险值对应的第一损失值及与所述样本关联风险值对应的第二损失值；第一参数值更新单元，用于根据所述梯度计算公式、所述第一损失值及所述病患风险预测模型的计算值计算得到所述病患风险预测模型中每一参数的 更新值以更新所述参数的参数值；第二参数值更新单元，用于根据所述梯度计算公式、所述第二损失值及所述关联风险预测模型的计算值计算得到所述关联风险预测模型中每一参数的更新值以更新所述参数的参数值。
关联风险值获取单元140，用于根据预存的关联风险预测模型获取与所述关联量化信息对应的关联风险值。
风险预测结果获取单元150，用于根据预存的风险系数计算公式对所述病患风险值及所述关联风险值进行计算以得到对应的风险预测结果并反馈至所述客户端。
在一实施例中，所述基于风险预测的就诊分配装置100还包括子单元：权重系数配置单元。
权重系数配置单元，用于根据所述预测类型获取预存的权重信息表中与所述预测类型对应的一组权重值，对所述风险系数计算公式中的权重系数进行配置。
就诊分配信息获取单元160，用于若接收到来自所述客户端的就诊请求，根据预存的当前就诊信息获取与所述风险预测结果及所述预测类型对应的就诊分配信息并反馈至所述客户端。
在一实施例中，所述就诊分配信息获取单元160包括子单元：判断单元及空闲诊区获取单元。
判断单元，用于对所述预测类型及所述风险预测结果是否满足预置的优先分配条件进行判断以得到判断结果；空闲诊区获取单元，用于若所述判断结果为是，获取所述当前就诊信息中与所述预测请求信息的个人信息对应的空闲诊区作为所述就诊分配信息反馈至所述客户端。
在本申请实施例所提供的基于风险预测的就诊分配装置应用上述基于风险预测的就诊分配方法，根据来自客户端的预测请求信息发送对应的询问信息至该客户端，并获取反馈的答复信息，根据信息量化规则对答复信息及预测请求信息中的病患信息进行量化得到对应的病患量化信息，对其中的关联信息进行量化得到对应的关联信息，根据病患风险预测模型获取与病患量化信息对应的病患风险值、根据关联风险预测模型获取与关联风险对应的关联风险值后进行综合得到风险预测结果，根据风险预测结果及预测类型获取对应的就诊分配信息并反馈至该客户端。通过上述方法，可依据病患风险预测模型获取对应的病患风险值、根据关联风险预测模型获取对应的关联风险值，通过两方面的信息对客户的患病风险更有针对性地进行评价，可更加全面地对客户的患病风险进行评估，并提高所得到的风险预测结果的准确性，从而提高了对患者进行就诊分配的效率及准确率。
上述基于风险预测的就诊分配装置可以实现为计算机程序的形式，该计算机程序可以在如图10所示的计算机设备上运行。
请参阅图10，图10是本申请实施例提供的计算机设备的示意性框图。该计算机设备可以是即是用于执行基于风险预测的就诊分配方法以对客户进行就诊分配的企业终端，例如，医院内所设置的服务器、健康服务企业所设置的服务器等。
参阅图10，该计算机设备500包括通过系统总线501连接的处理器502、存储器和网络接口505，其中，存储器可以包括非易失性存储介质503和内存储器504。
该非易失性存储介质503可存储操作系统5031和计算机程序5032。该计算机程序5032被执行时，可使得处理器502执行基于风险预测的就诊分配方法。
该处理器502用于提供计算和控制能力，支撑整个计算机设备500的运行。
该内存储器504为非易失性存储介质503中的计算机程序5032的运行提供环境，该计算机程序5032被处理器502执行时，可使得处理器502执行基于风险预测的就诊分配方法。
该网络接口505用于进行网络通信，如提供数据信息的传输等。本领域技术人员可以理解，图10中示出的结构，仅仅是与本申请方案相关的部分结构的框图，并不构成对本申请方 案所应用于其上的计算机设备500的限定，具体的计算机设备500可以包括比图中所示更多或更少的部件，或者组合某些部件，或者具有不同的部件布置。
其中，所述处理器502用于运行存储在存储器中的计算机程序5032，以实现上述的基于风险预测的就诊分配方法中对应的功能。
本领域技术人员可以理解，图10中示出的计算机设备的实施例并不构成对计算机设备具体构成的限定，在其他实施例中，计算机设备可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置。例如，在一些实施例中，计算机设备可以仅包括存储器及处理器，在这样的实施例中，存储器及处理器的结构及功能与图10所示实施例一致，在此不再赘述。
应当理解，在本申请实施例中，处理器502可以是中央处理单元(Central Processing Unit，CPU)，该处理器502还可以是其他通用处理器、数字信号处理器(Digital Signal Processor，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现成可编程门阵列(Field-Programmable Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。其中，通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
在本申请的另一实施例中提供计算机可读存储介质。该计算机可读存储介质可以为非易失性的计算机可读存储介质，也可以是易失性的计算机可读存储介质。该计算机可读存储介质存储有计算机程序，其中计算机程序被处理器执行时实现上述的基于风险预测的就诊分配方法中所包含的步骤。
所属领域的技术人员可以清楚地了解到，为了描述的方便和简洁，上述描述的设备、装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、计算机软件或者二者的结合来实现，为了清楚地说明硬件和软件的可互换性，在上述说明中已经按照功能一般性地描述了各示例的组成及步骤。这些功能究竟以硬件还是软件方式来执行取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
在本申请所提供的几个实施例中，应该理解到，所揭露的设备、装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为逻辑功能划分，实际实现时可以有另外的划分方式，也可以将具有相同功能的单元集合成一个单元，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另外，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口、装置或单元的间接耦合或通信连接，也可以是电的，机械的或其它的形式连接。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本申请实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以是两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。
所述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分，或者该技术方案的全部或部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个计算机可读存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的计算机可读存储介质包括：U盘、移动硬盘、只读存储器(ROM，Read-Only  Memory)、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述，仅为本申请的具体实施方式，但本申请的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，可轻易想到各种等效的修改或替换，这些修改或替换都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应以权利要求的保护范围为准。

Claims (20)

  1. 一种基于风险预测的就诊分配方法，应用于管理服务器，所述管理服务器与至少一台客户端进行通信，其中，所述方法包括：
    若接收到来自所述客户端的预测请求信息，根据预存的询问数据库及所述预测请求信息的预测类型获取与所述预测类型对应的询问信息并发送至所述客户端；
    若接收到所述客户端根据所述询问信息反馈的答复信息，根据预存的信息量化规则对所述答复信息及所述预测请求信息进行量化以得到对应的量化信息；其中，所述量化信息包括病患量化信息及关联量化信息；
    根据预存的病患风险预测模型获取与所述病患量化信息对应的病患风险值；
    根据预存的关联风险预测模型获取与所述关联量化信息对应的关联风险值；
    根据预存的风险系数计算公式对所述病患风险值及所述关联风险值进行计算以得到对应的风险预测结果并反馈至所述客户端；
    若接收到来自所述客户端的就诊请求，根据预存的当前就诊信息获取与所述风险预测结果及所述预测类型对应的就诊分配信息并反馈至所述客户端。
  2. 根据权利要求1所述的基于风险预测的就诊分配方法，其中，所述根据预存的询问数据库及所述预测请求信息的预测类型获取与所述预测类型对应的询问信息并发送至所述客户端，包括：
    获取所述询问数据库中与所述预测类型相匹配的第一询问数据；
    将所述询问数据库中的默认询问数据及所述第一询问数据组合为所述询问信息并发送至所述客户端。
  3. 根据权利要求1所述的基于风险预测的就诊分配方法，其中，所述信息量化规则包括病患项目信息、关联项目信息及项目值量化规则，所述根据预存的信息量化规则对所述答复信息及所述预测请求信息进行量化以得到对应的量化信息，包括：
    根据所述病患项目信息及所述关联项目信息对所述答复信息及所述预测请求信息中的个人信息进行分类，以得到与所述病患项目信息对应的病患信息及与所述关联项目信息对应的关联信息；
    根据所述项目值量化规则对所述病患信息的项目值及所述关联信息的项目值分别进行以得到对应的病患量化信息及关联量化信息。
  4. 根据权利要求1所述的基于风险预测的就诊分配方法，其中，所述根据预存的病患风险预测模型获取与所述病患量化信息对应的病患风险值之前，还包括：
    若接收到所述管理服务器的管理员所输入的样本数据库，根据所述样本数据库所包含的样本数据及预存的梯度计算公式分别对所述病患风险预测模型及所述关联风险预测模型进行训练，以得到训练后的所述病患风险预测模型及训练后的所述关联风险预测模型。
  5. 根据权利要求4所述的基于风险预测的就诊分配方法，其中，所述根据所述样本数据库所包含的样本数据及预存的梯度计算公式分别对所述病患风险预测模型及所述关联风险预测模型进行训练，包括：
    根据所述信息量化规则对一条所述样本数据进行量化以得到对应的样本量化信息，其中，所述样本量化信息包括样本诊断量化信息、样本病患量化信息及样本关联量化信息；
    根据所述病患风险预测模型获取与所述样本病患量化信息对应的样本病患风险值；
    根据所述关联风险预测模型获取与所述样本关联量化信息对应的样本关联风险值；
    根据所述样本诊断量化信息分别获取与所述样本病患风险值对应的第一损失值及与所述样本关联风险值对应的第二损失值；
    根据所述梯度计算公式、所述第一损失值及所述病患风险预测模型的计算值计算得到所 述病患风险预测模型中每一参数的更新值以更新所述参数的参数值；
    根据所述梯度计算公式、所述第二损失值及所述关联风险预测模型的计算值计算得到所述关联风险预测模型中每一参数的更新值以更新所述参数的参数值。
  6. 根据权利要求1所述的基于风险预测的就诊分配方法，其中，所述根据预存的风险系数计算公式对所述病患风险值及所述关联风险值进行计算以得到对应的风险预测结果并反馈至所述客户端之前，包括：
    根据所述预测类型获取预存的权重信息表中与所述预测类型对应的一组权重值，对所述风险系数计算公式中的权重系数进行配置。
  7. 根据权利要求1所述的基于风险预测的就诊分配方法，其中，所述根据预存的当前就诊信息获取与所述风险预测结果及所述预测类型对应的就诊分配信息并反馈至所述客户端，包括：
    对所述预测类型及所述风险预测结果是否满足预置的优先分配条件进行判断以得到判断结果；
    若所述判断结果为是，获取所述当前就诊信息中与所述预测请求信息的个人信息对应的空闲诊区作为所述就诊分配信息反馈至所述客户端。
  8. 一种基于风险预测的就诊分配装置，包括：
    询问信息发送单元，用于若接收到来自所述客户端的预测请求信息，根据预存的询问数据库及所述预测请求信息的预测类型获取与所述预测类型对应的询问信息并发送至所述客户端；
    信息量化单元，用于若接收到所述客户端根据所述询问信息反馈的答复信息，根据预存的信息量化规则对所述答复信息及所述预测请求信息进行量化以得到对应的量化信息；其中，所述量化信息包括病患量化信息及关联量化信息；
    病患风险值获取单元，用于根据预存的病患风险预测模型获取与所述病患量化信息对应的病患风险值；
    关联风险值获取单元，用于根据预存的关联风险预测模型获取与所述关联量化信息对应的关联风险值；
    风险预测结果获取单元，用于根据预存的风险系数计算公式对所述病患风险值及所述关联风险值进行计算以得到对应的风险预测结果并反馈至所述客户端；
    就诊分配信息获取单元，用于若接收到来自所述客户端的就诊请求，根据预存的当前就诊信息获取与所述风险预测结果及所述预测类型对应的就诊分配信息并反馈至所述客户端。
  9. 一种计算机设备，包括存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序，其中，所述处理器执行所述计算机程序时实现以下步骤：
    若接收到来自所述客户端的预测请求信息，根据预存的询问数据库及所述预测请求信息的预测类型获取与所述预测类型对应的询问信息并发送至所述客户端；
    若接收到所述客户端根据所述询问信息反馈的答复信息，根据预存的信息量化规则对所述答复信息及所述预测请求信息进行量化以得到对应的量化信息；其中，所述量化信息包括病患量化信息及关联量化信息；
    根据预存的病患风险预测模型获取与所述病患量化信息对应的病患风险值；
    根据预存的关联风险预测模型获取与所述关联量化信息对应的关联风险值；
    根据预存的风险系数计算公式对所述病患风险值及所述关联风险值进行计算以得到对应的风险预测结果并反馈至所述客户端；
    若接收到来自所述客户端的就诊请求，根据预存的当前就诊信息获取与所述风险预测结果及所述预测类型对应的就诊分配信息并反馈至所述客户端。
  10. 根据权利要求9所述的计算机设备，其中，所述根据预存的询问数据库及所述预测 请求信息的预测类型获取与所述预测类型对应的询问信息并发送至所述客户端，包括：
    获取所述询问数据库中与所述预测类型相匹配的第一询问数据；
    将所述询问数据库中的默认询问数据及所述第一询问数据组合为所述询问信息并发送至所述客户端。
  11. 根据权利要求9所述的计算机设备，其中，所述信息量化规则包括病患项目信息、关联项目信息及项目值量化规则，所述根据预存的信息量化规则对所述答复信息及所述预测请求信息进行量化以得到对应的量化信息，包括：
    根据所述病患项目信息及所述关联项目信息对所述答复信息及所述预测请求信息中的个人信息进行分类，以得到与所述病患项目信息对应的病患信息及与所述关联项目信息对应的关联信息；
    根据所述项目值量化规则对所述病患信息的项目值及所述关联信息的项目值分别进行以得到对应的病患量化信息及关联量化信息。
  12. 根据权利要求9所述的计算机设备，其中，所述根据预存的病患风险预测模型获取与所述病患量化信息对应的病患风险值之前，还包括：
    若接收到所述管理服务器的管理员所输入的样本数据库，根据所述样本数据库所包含的样本数据及预存的梯度计算公式分别对所述病患风险预测模型及所述关联风险预测模型进行训练，以得到训练后的所述病患风险预测模型及训练后的所述关联风险预测模型。
  13. 根据权利要求12所述的计算机设备，其中，所述根据所述样本数据库所包含的样本数据及预存的梯度计算公式分别对所述病患风险预测模型及所述关联风险预测模型进行训练，包括：
    根据所述信息量化规则对一条所述样本数据进行量化以得到对应的样本量化信息，其中，所述样本量化信息包括样本诊断量化信息、样本病患量化信息及样本关联量化信息；
    根据所述病患风险预测模型获取与所述样本病患量化信息对应的样本病患风险值；
    根据所述关联风险预测模型获取与所述样本关联量化信息对应的样本关联风险值；
    根据所述样本诊断量化信息分别获取与所述样本病患风险值对应的第一损失值及与所述样本关联风险值对应的第二损失值；
    根据所述梯度计算公式、所述第一损失值及所述病患风险预测模型的计算值计算得到所述病患风险预测模型中每一参数的更新值以更新所述参数的参数值；
    根据所述梯度计算公式、所述第二损失值及所述关联风险预测模型的计算值计算得到所述关联风险预测模型中每一参数的更新值以更新所述参数的参数值。
  14. 根据权利要求9所述的计算机设备，其中，所述根据预存的风险系数计算公式对所述病患风险值及所述关联风险值进行计算以得到对应的风险预测结果并反馈至所述客户端之前，包括：
    根据所述预测类型获取预存的权重信息表中与所述预测类型对应的一组权重值，对所述风险系数计算公式中的权重系数进行配置。
  15. 根据权利要求9所述的计算机设备，其中，所述根据预存的当前就诊信息获取与所述风险预测结果及所述预测类型对应的就诊分配信息并反馈至所述客户端，包括：
    对所述预测类型及所述风险预测结果是否满足预置的优先分配条件进行判断以得到判断结果；
    若所述判断结果为是，获取所述当前就诊信息中与所述预测请求信息的个人信息对应的空闲诊区作为所述就诊分配信息反馈至所述客户端。
  16. 一种计算机可读存储介质，其中，所述计算机可读存储介质存储有计算机程序，所述计算机程序当被处理器执行时使所述处理器执行以下操作：
    若接收到来自所述客户端的预测请求信息，根据预存的询问数据库及所述预测请求信息 的预测类型获取与所述预测类型对应的询问信息并发送至所述客户端；
    若接收到所述客户端根据所述询问信息反馈的答复信息，根据预存的信息量化规则对所述答复信息及所述预测请求信息进行量化以得到对应的量化信息；其中，所述量化信息包括病患量化信息及关联量化信息；
    根据预存的病患风险预测模型获取与所述病患量化信息对应的病患风险值；
    根据预存的关联风险预测模型获取与所述关联量化信息对应的关联风险值；
    根据预存的风险系数计算公式对所述病患风险值及所述关联风险值进行计算以得到对应的风险预测结果并反馈至所述客户端；
    若接收到来自所述客户端的就诊请求，根据预存的当前就诊信息获取与所述风险预测结果及所述预测类型对应的就诊分配信息并反馈至所述客户端。
  17. 根据权利要求16所述的计算机可读存储介质，其中，所述根据预存的询问数据库及所述预测请求信息的预测类型获取与所述预测类型对应的询问信息并发送至所述客户端，包括：
    获取所述询问数据库中与所述预测类型相匹配的第一询问数据；
    将所述询问数据库中的默认询问数据及所述第一询问数据组合为所述询问信息并发送至所述客户端。
  18. 根据权利要求16所述的计算机可读存储介质，其中，所述信息量化规则包括病患项目信息、关联项目信息及项目值量化规则，所述根据预存的信息量化规则对所述答复信息及所述预测请求信息进行量化以得到对应的量化信息，包括：
    根据所述病患项目信息及所述关联项目信息对所述答复信息及所述预测请求信息中的个人信息进行分类，以得到与所述病患项目信息对应的病患信息及与所述关联项目信息对应的关联信息；
    根据所述项目值量化规则对所述病患信息的项目值及所述关联信息的项目值分别进行以得到对应的病患量化信息及关联量化信息。
  19. 根据权利要求16所述的计算机可读存储介质，其中，所述根据预存的病患风险预测模型获取与所述病患量化信息对应的病患风险值之前，还包括：
    若接收到所述管理服务器的管理员所输入的样本数据库，根据所述样本数据库所包含的样本数据及预存的梯度计算公式分别对所述病患风险预测模型及所述关联风险预测模型进行训练，以得到训练后的所述病患风险预测模型及训练后的所述关联风险预测模型。
  20. 根据权利要求19所述的计算机可读存储介质，其中，所述根据所述样本数据库所包含的样本数据及预存的梯度计算公式分别对所述病患风险预测模型及所述关联风险预测模型进行训练，包括：
    根据所述信息量化规则对一条所述样本数据进行量化以得到对应的样本量化信息，其中，所述样本量化信息包括样本诊断量化信息、样本病患量化信息及样本关联量化信息；
    根据所述病患风险预测模型获取与所述样本病患量化信息对应的样本病患风险值；
    根据所述关联风险预测模型获取与所述样本关联量化信息对应的样本关联风险值；
    根据所述样本诊断量化信息分别获取与所述样本病患风险值对应的第一损失值及与所述样本关联风险值对应的第二损失值；
    根据所述梯度计算公式、所述第一损失值及所述病患风险预测模型的计算值计算得到所述病患风险预测模型中每一参数的更新值以更新所述参数的参数值；
    根据所述梯度计算公式、所述第二损失值及所述关联风险预测模型的计算值计算得到所述关联风险预测模型中每一参数的更新值以更新所述参数的参数值。
PCT/CN2021/097209 2020-07-16 2021-05-31 基于风险预测的就诊分配方法、装置、计算机设备 Ceased WO2022012188A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010688314.0A CN111833997B (zh) 2020-07-16 2020-07-16 基于风险预测的就诊分配方法、装置、计算机设备
CN202010688314.0 2020-07-16

Publications (1)

Publication Number Publication Date
WO2022012188A1 true WO2022012188A1 (zh) 2022-01-20

Family

ID=72924328

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/097209 Ceased WO2022012188A1 (zh) 2020-07-16 2021-05-31 基于风险预测的就诊分配方法、装置、计算机设备

Country Status (2)

Country Link
CN (1) CN111833997B (zh)
WO (1) WO2022012188A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111833997B (zh) * 2020-07-16 2023-05-30 平安科技（深圳）有限公司 基于风险预测的就诊分配方法、装置、计算机设备
CN112530598B (zh) * 2020-12-11 2023-07-25 万达信息股份有限公司 一种基于健康数据的健康风险自测量表推荐方法
CN114678129A (zh) * 2022-03-28 2022-06-28 合和公卫(广州)安全科技有限公司 一种环境健康码的构建方法与平台

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150213224A1 (en) * 2012-09-13 2015-07-30 Parkland Center For Clinical Innovation Holistic hospital patient care and management system and method for automated patient monitoring
US20170177824A1 (en) * 2015-12-18 2017-06-22 Medical Home Network Healthcare management system and method for evaluating patients
US20170286622A1 (en) * 2016-03-29 2017-10-05 International Business Machines Corporation Patient Risk Assessment Based on Machine Learning of Health Risks of Patient Population
CN108784655A (zh) * 2017-04-28 2018-11-13 西门子保健有限责任公司 针对医疗患者的快速评估和后果分析
CN111833997A (zh) * 2020-07-16 2020-10-27 平安科技（深圳）有限公司 基于风险预测的就诊分配方法、装置、计算机设备

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9465917B2 (en) * 2014-05-30 2016-10-11 Roche Diabetes Care, Inc. Hazard based assessment patterns
EP4437962A3 (en) * 2015-12-18 2025-01-01 Cognoa, Inc. Platform and system for digital personalized medicine
CN108122612A (zh) * 2017-12-20 2018-06-05 姜涵予 数据库的建立、多维度健康风险等级确定方法及装置
CN108766568A (zh) * 2018-04-19 2018-11-06 新绎健康科技（北京）有限公司 一种疾病风险自动预警方法及系统
CN109119130A (zh) * 2018-07-11 2019-01-01 上海夏先机电科技发展有限公司 一种基于云计算的大数据健康管理系统及方法
CN109003193B (zh) * 2018-09-10 2024-06-28 平安科技（深圳）有限公司 投保风险预测方法、装置、计算机设备和存储介质
CN109065139A (zh) * 2018-09-10 2018-12-21 平安科技（深圳）有限公司 医疗随访方法、装置、计算机设备和存储介质
CN109859847A (zh) * 2019-02-15 2019-06-07 京东方科技集团股份有限公司 电子设备、体重管理收益预测装置及存储介质
CN110111886A (zh) * 2019-05-16 2019-08-09 闻康集团股份有限公司 一种基于XGBoost疾病预测的智能问诊系统及方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150213224A1 (en) * 2012-09-13 2015-07-30 Parkland Center For Clinical Innovation Holistic hospital patient care and management system and method for automated patient monitoring
US20170177824A1 (en) * 2015-12-18 2017-06-22 Medical Home Network Healthcare management system and method for evaluating patients
US20170286622A1 (en) * 2016-03-29 2017-10-05 International Business Machines Corporation Patient Risk Assessment Based on Machine Learning of Health Risks of Patient Population
CN108784655A (zh) * 2017-04-28 2018-11-13 西门子保健有限责任公司 针对医疗患者的快速评估和后果分析
CN111833997A (zh) * 2020-07-16 2020-10-27 平安科技（深圳）有限公司 基于风险预测的就诊分配方法、装置、计算机设备

Also Published As

Publication number Publication date
CN111833997A (zh) 2020-10-27
CN111833997B (zh) 2023-05-30

Similar Documents

Publication Publication Date Title
US11600390B2 (en) Machine learning clinical decision support system for risk categorization
US10217048B2 (en) Predictive assignments that relate to genetic information and leverage machine learning models
US10790048B2 (en) Patient treatment recommendations based on medical records and exogenous information
US10565309B2 (en) Interpreting the meaning of clinical values in electronic medical records
US20170286622A1 (en) Patient Risk Assessment Based on Machine Learning of Health Risks of Patient Population
US20170293733A1 (en) Dynamic Selection and Sequencing of Healthcare Assessments for Patients
CN112117011A (zh) 基于人工智能的传染病早期风险预警方法、装置
US20240112794A1 (en) Time-based resource allocation for long-term integrated health computer system
WO2022012188A1 (zh) 基于风险预测的就诊分配方法、装置、计算机设备
Hameed Abdulkareem et al. [Retracted] Smart Healthcare System for Severity Prediction and Critical Tasks Management of COVID‐19 Patients in IoT‐Fog Computing Environments
US11710572B2 (en) Experience engine-method and apparatus of learning from similar patients
US11699069B2 (en) Predictive assignments that relate to genetic information and leverage machine learning models
McCombs et al. A model-based evaluation of the efficacy of COVID-19 social distancing, testing and hospital triage policies
CN104881438A (zh) 网络医院平台、专家平台及基于专家平台的紧急专家会诊请求方法
CN104866711A (zh) 用户终端、专家平台及基于用户终端的专家会诊请求方法
WO2021211326A1 (en) Systems and methods for access management and clustering of genomic, phenotype, and diagnostic data
JP6726273B2 (ja) データ品質管理システム及び方法
JP2019530098A (ja) 協調的な変異選択及び治療合致レポートのための方法及び装置
CN112750529A (zh) 智能医疗问诊装置、设备及介质
Pahlevani et al. Developing a decision support tool to predict delayed discharge from hospitals using machine learning
Krieger et al. A blueprint for identifying phenotypes and drug targets in complex disorders with empirical dynamics
JP7772539B2 (ja) 情報処理装置、情報処理システム、情報処理方法およびプログラム
CN112542219B (zh) 患者入院管理方法、装置、管理设备及可读存储介质
Chen et al. Identifying low acuity Emergency Department visits with a machine learning approach: The low acuity visit algorithms (LAVA)
CN119993530A (zh) 一种基于大数据的疾病咨询系统

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21841207

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21841207

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 25/04/2023)

122 Ep: pct application non-entry in european phase

Ref document number: 21841207

Country of ref document: EP

Kind code of ref document: A1