WO2021203997A1 - Procédé de prévision de coûts d'assurance maladie associés à une maladie chronique en fonction du risque de complications, et dispositif associé - Google Patents

Procédé de prévision de coûts d'assurance maladie associés à une maladie chronique en fonction du risque de complications, et dispositif associé Download PDF

Info

Publication number
WO2021203997A1
WO2021203997A1 PCT/CN2021/083523 CN2021083523W WO2021203997A1 WO 2021203997 A1 WO2021203997 A1 WO 2021203997A1 CN 2021083523 W CN2021083523 W CN 2021083523W WO 2021203997 A1 WO2021203997 A1 WO 2021203997A1
Authority
WO
WIPO (PCT)
Prior art keywords
chronic disease
data
historical
complication
disease patient
Prior art date
Application number
PCT/CN2021/083523
Other languages
English (en)
Chinese (zh)
Inventor
徐啸
徐衔
孙瑜尧
李响
谢国彤
Original Assignee
平安科技（深圳）有限公司
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 平安科技（深圳）有限公司 filed Critical 平安科技（深圳）有限公司
Publication of WO2021203997A1 publication Critical patent/WO2021203997A1/fr

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
    • 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
    • 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
    • G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
    • G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
    • 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

Definitions

  • This application relates to the field of medical technology, and in particular to a method and related equipment for predicting the cost of chronic disease medical insurance for the risk of fusion complications.
  • the embodiment of the application provides a method and related equipment for predicting the cost of chronic disease medical insurance with the risk of fusion complications, which can predict the risk probability of chronic disease patients suffering from complications, and integrate the risk probability of chronic disease patients suffering from complications into the medical insurance cost prediction In the model, it helps to improve the accuracy of predicting the medical insurance costs of patients with chronic diseases.
  • an embodiment of the present application provides a device for predicting medical insurance costs for chronic diseases with a risk of fusion complications, the device including: a memory and a processor;
  • the memory is used to store program instructions
  • the processor is configured to call the program instructions, and when the program instructions are executed, to perform the following operations:
  • the chronic disease population data set including historical consultation data, historical medical insurance cost outcome data, and historical complications outcome data of multiple chronic disease patients, the historical consultation data including historical chronic disease data and historical concurrency Symptom data, wherein the historical chronic disease data is diagnostic data used to indicate chronic disease, and the historical complication data is diagnostic data used to indicate complications;
  • the graph network structure corresponding to the medical knowledge graph of the historical complication data, and determine the characterization of each node in the graph network structure
  • a vector wherein the graph network structure is composed of multiple nodes and edges, each node is a complication, and each edge refers to the correlation between two complication;
  • the historical chronic disease data of each chronic disease patient and the probability of each chronic disease patient suffering from each complication are input into the first recurrent neural network model to obtain the medical insurance cost result of each chronic disease patient, and according to Training the first recurrent neural network model on the result of the medical insurance cost of each chronic disease patient to obtain a medical insurance cost prediction model;
  • the embodiment of the present application provides a method for predicting the cost of chronic disease medical insurance for the risk of fusion complications, including:
  • the chronic disease population data set including historical consultation data, historical medical insurance cost outcome data, and historical complications outcome data of multiple chronic disease patients, the historical consultation data including historical chronic disease data and historical concurrency Symptom data, wherein the historical chronic disease data is diagnostic data used to indicate chronic disease, and the historical complication data is diagnostic data used to indicate complications;
  • the graph network structure corresponding to the medical knowledge graph of the historical complication data, and determine the characterization of each node in the graph network structure
  • a vector wherein the graph network structure is composed of multiple nodes and edges, each node is a complication, and each edge refers to the correlation between two complication;
  • the historical chronic disease data of each chronic disease patient and the probability of each chronic disease patient suffering from each complication are input into the first recurrent neural network model to obtain the medical insurance cost result of each chronic disease patient, and according to Training the first recurrent neural network model on the result of the medical insurance cost of each chronic disease patient to obtain a medical insurance cost prediction model;
  • an embodiment of the present application provides a device for predicting the cost of chronic disease medical insurance for the risk of fusion complications, including:
  • the acquiring unit is configured to acquire a chronic disease population data set, the chronic disease population data set including historical consultation data, historical medical insurance cost outcome data, and historical complications outcome data of a plurality of chronic disease patients, the historical consultation data including historical chronic illness Disease data and historical complication data, wherein the historical chronic disease data is diagnostic data used to indicate a chronic disease, and the historical complication data is diagnostic data used to indicate a complication;
  • the first determining unit is configured to determine each chronic disease data set in the chronic disease population data set according to the historical chronic disease data and the historical complication data in the historical consultation data of each chronic disease patient in the chronic disease population data set The frequency of co-occurrence among different complications in patients with chronic diseases;
  • the second determining unit is configured to determine the graph network structure corresponding to the medical knowledge graph of the historical complication data according to the frequency of co-occurrence between the different complications of each chronic disease patient, and determine the graph network
  • the first training unit is used to input the historical chronic disease data of each chronic disease patient in the chronic disease population data set and the representation vector of each node in the graph network structure into the multi-layer perceptron model to obtain the chronic disease population
  • the second training unit is used to input the historical chronic disease data of each chronic disease patient and the probability of each chronic disease patient suffering from each complication into the first recurrent neural network model to obtain each chronic disease patient And training the first recurrent neural network model to obtain a medical insurance cost prediction model according to the medical insurance cost results of each chronic disease patient;
  • the prediction unit is used to obtain the medical treatment data of the chronic disease patient to be tested, and input the medical consultation data of the chronic disease patient to be tested into the complication prediction model and the medical insurance cost prediction model to obtain the risk of complications of the chronic disease patient to be tested Probability and medical insurance cost forecast results.
  • an embodiment of the present application provides a computer-readable storage medium, the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the following method:
  • the chronic disease population data set including historical consultation data, historical medical insurance cost outcome data, and historical complications outcome data of multiple chronic disease patients, the historical consultation data including historical chronic disease data and historical concurrency Symptom data, wherein the historical chronic disease data is diagnostic data used to indicate chronic disease, and the historical complication data is diagnostic data used to indicate complications;
  • the graph network structure corresponding to the medical knowledge graph of the historical complication data, and determine the characterization of each node in the graph network structure
  • a vector wherein the graph network structure is composed of multiple nodes and edges, each node is a complication, and each edge refers to the correlation between two complication;
  • the historical chronic disease data of each chronic disease patient and the probability of each chronic disease patient suffering from each complication are input into the first recurrent neural network model to obtain the medical insurance cost result of each chronic disease patient, and according to Training the first recurrent neural network model on the result of the medical insurance cost of each chronic disease patient to obtain a medical insurance cost prediction model;
  • the embodiments of the present application can predict the risk probability of chronic disease patients suffering from complications, and integrate the risk probability of chronic disease patients suffering complications into the medical insurance cost prediction model, which helps to improve the accuracy of predicting the medical insurance costs of chronic disease patients.
  • Fig. 1 is a schematic structural diagram of a medical insurance cost prediction system provided by an embodiment of the present application
  • FIG. 2 is a schematic flowchart of a method for predicting medical insurance costs for chronic diseases of the risk of fusion complications according to an embodiment of the present application;
  • FIG. 3 is a schematic block diagram of a device for predicting the cost of chronic disease medical insurance for the risk of fusion complications according to an embodiment of the present application
  • Fig. 4 is a schematic block diagram of a device for predicting medical insurance costs for chronic diseases with a risk of fusion complications according to an embodiment of the present application.
  • the technical solution of this application relates to the field of artificial intelligence technology, and can be applied to scenarios such as smart medical care to realize digital medical care and promote the construction of smart cities.
  • the data involved in this application can be stored in a database, or can be stored in a blockchain, which is not limited by this application.
  • the method for predicting the cost of chronic disease medical insurance for the risk of fusion complications can be applied to a medical insurance cost prediction system.
  • the medical insurance cost prediction system includes a medical server and the risk of fusion complications Equipment for predicting medical insurance costs.
  • the medical server may establish a communication connection with a chronic disease medical insurance cost prediction device that integrates the risk of complications.
  • the communication connection mode may include, but is not limited to, Wi-Fi, Bluetooth, Near Field Communication (NFC), and so on.
  • the medical server is used to store a data set of people with chronic diseases.
  • the medical insurance cost prediction system provided by the embodiment of the present application will be schematically described below with reference to FIG. 1.
  • FIG. 1 is a schematic structural diagram of a medical insurance cost prediction system provided by an embodiment of the present application.
  • the medical insurance cost prediction system includes: a chronic disease medical insurance cost prediction device 11 and a medical server 12 that integrate the risk of complications.
  • the chronic disease medical insurance cost prediction device 11 and the medical server 12 for the risk of fusion complications may establish a communication connection through a wireless communication connection; wherein, in some scenarios, the chronic disease risk of the fusion complications
  • the medical insurance cost prediction device 11 and the medical server 12 may also establish a communication connection through a wired communication connection.
  • the device 11 for predicting the cost of chronic disease medical insurance for the risk of fusion complications may include, but is not limited to, smart terminal devices such as smart phones, tablet computers, notebook computers, and desktop computers.
  • the chronic disease medical insurance cost prediction device 11 that integrates the risk of complications can obtain a chronic disease population data set from the medical server 12, and the chronic disease population data set includes historical visit data and history of multiple chronic disease patients. Outcome data of medical insurance expenses and outcome data of historical complications, the historical visit data includes historical chronic disease data and historical complications data, wherein the historical chronic disease data is diagnostic data used to indicate a chronic disease, and the historical complications The data is diagnostic data used to indicate complications.
  • the chronic disease medical insurance cost prediction device 11 that integrates the risk of complications can determine the chronic disease data and the historical complication data in the historical visit data of each chronic disease patient in the chronic disease population data set.
  • the presentation vector of each chronic disease patient in the patient population data set input the presentation feature vector of each chronic disease patient into a designated classification model to obtain the probability of each chronic disease patient in the chronic patient cluster suffering from complications , And train the designated classification model according to the probability of each chronic disease patient suffering from a complication to obtain a complication prediction model;
  • the probability of the disease is input to the first recurrent neural network model to obtain the medical insurance cost result of each chronic disease patient, and the first recurrent neural network model is trained according to the medical insurance cost result of each chronic disease patient to obtain the medical insurance cost prediction Model;
  • heart failure and myocardial infarction are both macrovascular complications.
  • the occurrence of either heart failure or myocardial infarction will significantly increase the risk of the other. Therefore, through Introducing the correlation between complications when measuring the risk of complications can more accurately predict the risk probability of complications for patients with chronic diseases, and integrate the risk probability of complications for patients with chronic diseases into the medical insurance cost prediction model. To improve the accuracy of predicting the medical insurance costs of patients with chronic diseases.
  • FIG. 2 is a schematic flowchart of a method for predicting the cost of chronic disease medical insurance for the risk of fusion complications according to an embodiment of the present application. As shown in FIG. The cost prediction device is implemented, and the specific explanation of the chronic disease medical insurance cost prediction device for the risk of fusion complications is as described above, and will not be repeated here. Specifically, the method of the embodiment of the present application includes the following steps.
  • S201 Acquire a chronic disease population data set, where the chronic disease population data set includes historical visit data, historical medical insurance cost outcome data, and historical complications outcome data of multiple chronic disease patients, and the historical visit data includes historical chronic disease data and Historical complications data.
  • the chronic disease medical insurance cost prediction device integrating the risk of complications can obtain a chronic disease population data set, the chronic disease population data set including multiple chronic disease patients’ historical visit data, historical medical insurance cost outcome data and history Complication outcome data, the historical visit data includes historical chronic disease data and historical complication data, wherein the historical chronic disease data is diagnostic data used to indicate chronic disease, and the historical complication data is used to indicate concurrency Diagnosis data of the disease.
  • historical chronic disease data includes blood glucose data, urine protein data, etc., used to indicate diabetes
  • historical complications data includes vision data used to indicate cataracts, examination data used to indicate kidney disease, and the like.
  • S202 Determine the difference between each chronic disease patient in the chronic disease population data set according to the historical chronic disease data and the historical complication data in the historical visit data of each chronic disease patient in the chronic disease population data set Frequency of co-occurrence between complications.
  • the equipment for predicting the cost of chronic disease medical insurance that integrates the risk of complications may be based on the historical chronic disease data and the historical complications data in the historical visit data of each chronic disease patient in the chronic disease population data set To determine the frequency of co-occurrence among different complications of each chronic disease patient in the chronic disease population data set.
  • S203 Determine the graph network structure corresponding to the medical knowledge graph of the historical complication data according to the co-occurrence frequency between the different complications of each chronic disease patient, and determine each node in the graph network structure The representation vector.
  • the equipment for predicting the cost of chronic disease medical insurance for the risk of fusion complications can determine the medical knowledge map corresponding to the historical complication data according to the frequency of co-occurrence between different complications of each chronic disease patient And determine the representation vector of each node in the graph network structure, where the graph network structure is composed of multiple nodes and edges, each node is a complication, and each edge refers to two There is a correlation between the two complications.
  • the device for predicting the cost of chronic disease medical insurance fused with the risk of complications determines the characterization vector of each node in the graph network structure
  • it may perform a first encoding method on each chronic disease population data set in the chronic disease population data set.
  • the historical chronic disease data and the historical complication data in the patient’s historical consultation data are encoded to obtain the first encoded data;
  • the chronic disease population data is collected to correspond to the historical consultation data of each chronic patient
  • the first coded data is input into the second recurrent neural network model to obtain the medical representation vector of the historical visit data of each chronic patient in the chronic disease population data set, and the medical representation according to the historical visit data of each chronic patient
  • the vector determines the characterization vector of each node in the graph network structure.
  • the first encoding method may be a one-hot encoding method.
  • the equipment for predicting the cost of chronic disease medical insurance fusing the risk of complications inputs the first coded data corresponding to the historical visit data of each chronic patient in the chronic disease population data set into the second recurrent neural network model to obtain
  • the visit characterization vector of the historical visit data of each chronic patient in the chronic disease population data set the first encoded data corresponding to the historical visit data of each chronic patient patient in the chronic disease population data set may be input into the second loop
  • the neural network model obtains the chronic disease characterization vector corresponding to the historical chronic disease data of each chronic disease patient in the chronic disease population data set and the complication characterization vector corresponding to the historical complication data; and according to the chronic disease
  • the chronic disease characterization vector and the complication characterization vector of each chronic disease patient in the population data set are determined to determine the visit characterization vector of each chronic disease patient in the chronic disease population data set.
  • S204 Input the historical chronic disease data of each chronic disease patient in the chronic disease population data set and the representation vector of each node in the graph network structure into the multi-layer perceptron model to obtain each chronic disease population data set in the chronic disease population data set.
  • the probability of the patient suffering from each complication, and the multi-layer perceptron model is trained according to the probability of each complication of each chronic disease patient to obtain a complication prediction model.
  • the chronic disease medical insurance cost prediction device integrating the risk of complications can input the historical chronic disease data of each chronic disease patient in the chronic disease population data set and the characterization vector of each node in the graph network structure.
  • the multi-layer perceptron model is used to obtain the probability of each chronic disease patient suffering from each complication in the chronic disease population data set, and the multi-layer perceptron model is trained according to the probability of each chronic disease patient suffering from each complication to obtain the concurrency Disease prediction model.
  • the equipment for predicting the cost of chronic disease medical insurance for the risk of fusion complication trains the multi-layer perceptron model according to the probability of each chronic disease patient suffering from a complication to obtain a complication prediction model.
  • the probability of each chronic disease patient suffering from complications is compared with the historical complication outcome data of each chronic disease patient; the parameters of the multilayer perceptron model are adjusted according to the comparison result, and each chronic disease patient
  • the patient's diagnosis vector is input to the multi-layer perceptron model after adjusting the parameters for training, and the complication prediction model is obtained.
  • S205 Input the historical chronic disease data of each chronic disease patient and the probability of each chronic disease patient suffering from each complication into the first recurrent neural network model to obtain the medical insurance cost result of each chronic disease patient, And training the first recurrent neural network model according to the medical insurance cost result of each chronic disease patient to obtain a medical insurance cost prediction model.
  • the equipment for predicting the cost of chronic disease medical insurance for the risk of fusion complications can input the historical chronic disease data of each chronic disease patient and the probability of each chronic disease patient suffering from each complication into the first circulatory nerve.
  • the network model obtains the medical insurance cost result of each chronic disease patient, and trains the first recurrent neural network model according to the medical insurance cost result of each chronic disease patient to obtain a medical insurance cost prediction model.
  • the chronic disease medical insurance cost prediction device for fusion risk of complications inputs the historical chronic disease data of each chronic disease patient in the chronic disease population data set and the probability of each chronic disease patient suffering from complications.
  • the first cyclic neural network model when the medical insurance cost result of each chronic disease patient is obtained, the chronic disease characterization vector, the complication characterization vector and the complication characterization vector of each chronic disease patient can be collected in the data of the chronic disease population.
  • the probability of each chronic disease patient suffering from complications is input into the first recurrent neural network model to obtain the chronic disease medical insurance cost result corresponding to the historical chronic disease data of each chronic disease patient in the chronic disease population data set and each The probability of chronic disease patients suffering from complications corresponds to the complication medical insurance cost result; and according to the chronic disease medical insurance cost result of each chronic disease patient and the complication medical insurance cost result of each chronic disease patient, Determine the medical insurance cost result of each chronic disease patient.
  • a chronic disease medical insurance cost prediction device that integrates the risk of complications can input the historical chronic disease data of each diabetic patient and the probability of each diabetic patient in the chronic disease population data set into the first loop neural network model to obtain each Diabetes medical insurance cost results corresponding to the historical chronic disease data of diabetic patients and renal failure medical insurance cost results corresponding to the probability of each diabetic patient suffering from renal failure complications, and according to the results of diabetes medical insurance cost and renal failure medical insurance cost results of each diabetic patient , To determine the results of medical insurance costs for each diabetic patient.
  • the equipment for predicting the cost of medical insurance for chronic diseases with the risk of fusion complication trains the first recurrent neural network model according to the results of the medical insurance cost of each chronic disease patient to obtain the medical insurance cost prediction model
  • the medical insurance cost result of each chronic disease patient is compared with the historical medical insurance cost outcome data of each chronic disease patient; the parameters of the first recurrent neural network model are adjusted according to the comparison result, and each chronic disease patient is
  • the chronic disease characterization vector, the complication characterization vector, and the probability of each chronic disease patient suffering from a complication are inputted into a first recurrent neural network model after adjusting the parameters for training to obtain the medical insurance cost prediction model.
  • S206 Obtain the medical visit data of the chronically ill patient to be tested, and input the medical visit data of the chronically ill patient to be tested into the complication prediction model and the medical insurance cost prediction model to obtain the risk probability of complications and medical insurance for the chronically ill patient to be tested Cost forecast results.
  • the equipment for predicting medical insurance costs for chronic diseases that integrates the risk of complications can obtain medical treatment data of patients with chronic diseases to be tested, and input the medical care data of patients with chronic diseases to be tested into the complication prediction model and medical insurance cost predictions.
  • the model obtains the risk probability of complications and the prediction results of medical insurance costs for the patients with chronic diseases to be tested.
  • the equipment for predicting the cost of chronic disease medical insurance for the risk of fusion complication inputs the medical care data of the chronic disease patient to be tested into the complication prediction model and the medical insurance cost prediction model, and obtains that the patient with chronic disease to be tested
  • the chronic disease characterization vector and the complication characterization vector corresponding to the diagnosis data can be determined according to the diagnosis data of the chronic disease patient to be tested
  • the chronic disease characterization vector and the complication characterization vector are input into the complication prediction model to obtain the risk probability of the chronic disease patient to be tested for complications; and the risk probability of the chronic disease patient to be tested for the complications and the
  • the chronic disease characterization vector and the complication characterization vector of the chronic disease patient to be tested are input into the medical insurance cost prediction model to obtain the chronic disease medical insurance cost prediction result of the chronic disease patient to be tested and the complications of the chronic disease patient to be tested
  • the chronic disease medical insurance cost prediction device integrating the risk of complications can obtain a chronic disease population data set.
  • the chronic disease population data set includes historical visit data of multiple chronic disease patients, historical medical insurance cost outcome data, and historical complications
  • historical visit data includes historical chronic disease data and historical complications data; according to the historical chronic disease data and the historical complications data in the historical visit data of each chronic patient patient in the chronic disease population data set, Determine the frequency of co-occurrence between different complications of each chronic patient in the chronic disease population data set; determine the co-occurrence frequency between the different complications of each chronic patient
  • the graph network structure corresponding to the medical knowledge graph of the data, and the characterization vector of each node in the graph network structure is determined, where the graph network structure is composed of multiple nodes and edges, and each node is a complication, Each edge refers to the correlation between two complications; the historical chronic disease data of each chronic disease patient in the chronic disease population data set and the representation vector of each node in the graph network structure are input into the multi-layer perceptron model ,
  • FIG. 3 is a schematic block diagram of a device for predicting medical insurance costs for a chronic disease with a risk of fusion complications provided by an embodiment of the present application.
  • the device for predicting the cost of chronic disease medical insurance for the risk of fusion complications in this embodiment includes: an acquiring unit 301, a first determining unit 302, a second determining unit 303, a first training unit 304, a second training unit 305, and a predicting unit 306.
  • the acquiring unit 301 is configured to acquire a chronic disease population data set, the chronic disease population data set including historical consultation data, historical medical insurance cost outcome data, and historical complications outcome data of multiple chronic disease patients, the historical consultation data including history Chronic disease data and historical complication data, wherein the historical chronic disease data is diagnostic data used to indicate a chronic disease, and the historical complication data is diagnostic data used to indicate a complication;
  • the first determining unit 302 is configured to determine each chronic disease data set in the chronic disease population data set according to the historical chronic disease data and the historical complication data in the historical consultation data of each chronic disease patient in the chronic disease population data set. The frequency of co-occurrence among different complications of patients with chronic diseases;
  • the second determining unit 303 is configured to determine the graph network structure corresponding to the medical knowledge graph of the historical complication data according to the co-occurrence frequency between the different complications of each chronic disease patient, and determine the graph The representation vector of each node in the network structure, where the graph network structure is composed of multiple nodes and edges, each node is a complication, and each edge refers to the correlation between two complication;
  • the first training unit 304 is used to input the historical chronic disease data of each chronic disease patient in the chronic disease population data set and the representation vector of each node in the graph network structure into the multi-layer perceptron model to obtain the chronic disease
  • the second training unit 305 is used to input the historical chronic disease data of each chronic disease patient and the probability of each chronic disease patient suffering from each complication into the first recurrent neural network model to obtain each chronic disease
  • the prediction unit 306 is used to obtain the medical treatment data of the chronically ill patient to be tested, input the medical consultation data of the chronically ill patient to be tested into the complication prediction model and the medical insurance cost prediction model, and obtain the chronic disease patient's complication Risk probability and medical insurance cost prediction results.
  • the second determining unit 303 determines the characterization vector of each node in the graph network structure, it is specifically configured to:
  • the first coded data corresponding to the historical visit data of each chronic patient in the chronic disease population data set is input into the second cyclic neural network model to obtain the medical visit of the historical visit data of each chronic disease patient in the chronic disease population data set
  • a characterization vector, and the characterization vector of each node in the graph network structure is determined according to the medical characterization vector of the historical medical examination data of each chronic disease patient.
  • the second determining unit 303 inputs the first coded data corresponding to the historical visit data of each chronic disease patient in the chronic disease population data set into the second recurrent neural network model to obtain each chronic disease population data set.
  • the treatment representation vector of the historical treatment data of a chronic disease patient it is specifically used for:
  • the first coded data corresponding to the historical visit data of each chronic patient in the chronic disease population data set is input into the second cyclic neural network model to obtain the historical chronic disease of each chronic disease patient in the chronic disease population data set
  • the visit characterization vector of each chronic disease patient in the chronic disease population data set is determined.
  • the first training unit 304 trains the multilayer perceptron model according to the probability of each chronic disease patient suffering from a complication, and when a complication prediction model is obtained, it is specifically used for:
  • the second training unit 305 inputs the historical chronic disease data of each chronic disease patient in the chronic disease population data set and the probability of each chronic disease patient suffering from complications into the first recurrent neural network model to obtain When the result of the medical insurance cost of each chronic disease patient, it is specifically used for:
  • the chronic disease characterization vector, the complication characterization vector and the probability of each chronic disease patient suffering from the complications of each chronic disease patient in the chronic disease population data set are input into the first recurrent neural network model to obtain all State the chronic disease medical insurance cost result corresponding to the historical chronic disease data of each chronic disease patient in the chronic disease population data set and the complication medical insurance cost result corresponding to the probability of each chronic disease patient suffering from complications;
  • the medical insurance cost result of each chronic disease patient is determined according to the chronic disease medical insurance cost result of each chronic disease patient and the complication medical insurance cost result of each chronic disease patient.
  • the second training unit 305 trains the first recurrent neural network model to obtain a medical insurance cost prediction model according to the results of the medical insurance cost of each chronic disease patient, it is specifically used for:
  • the parameters of the first recurrent neural network model are adjusted according to the comparison result, and the chronic disease characterization vector, the complication characterization vector of each chronic disease patient, and the complication characterization vector of each chronic disease patient are adjusted. Probably input the first recurrent neural network model after adjusting the parameters for training, and obtain the medical insurance cost prediction model.
  • the predicting unit 306 inputs the visit data of the chronically ill patient to be tested into the complication prediction model and the medical insurance cost prediction model, and obtains the risk probability of the chronically ill patient to be tested and the medical insurance cost prediction result , Specifically used for:
  • the risk probability of the chronic disease patient to be tested and the chronic disease characterization vector and the complication characterization vector of the chronic disease patient to be tested are input into the medical insurance cost prediction model to obtain the chronic disease patient's chronic disease
  • FIG. 4 is a schematic block diagram of a device for predicting medical insurance costs for chronic diseases with a risk of fusion complications provided by an embodiment of the present application.
  • the device in this embodiment as shown in the figure may include: one or more processors 401 and a memory 402.
  • the memory 402 is configured to store a computer program, and the computer program includes a program, and the processor 401 is configured to execute the program stored in the memory 402.
  • the processor 401 is configured to call the program to execute:
  • the chronic disease population data set includes historical visit data, historical medical insurance cost outcome data, and historical complications outcome data of multiple chronic disease patients, and the historical visit data includes historical chronic disease data and historical concurrency Symptom data, wherein the historical chronic disease data is diagnostic data used to indicate chronic disease, and the historical complication data is diagnostic data used to indicate complications;
  • the graph network structure corresponding to the medical knowledge graph of the historical complication data, and determine the characterization of each node in the graph network structure
  • a vector wherein the graph network structure is composed of multiple nodes and edges, each node is a complication, and each edge refers to the correlation between two complication;
  • the historical chronic disease data of each chronic disease patient and the probability of each chronic disease patient suffering from each complication are input into the first recurrent neural network model to obtain the medical insurance cost result of each chronic disease patient, and according to Training the first recurrent neural network model on the result of the medical insurance cost of each chronic disease patient to obtain a medical insurance cost prediction model;
  • the processor 401 determines the characterization vector of each node in the graph network structure, it is specifically configured to:
  • the first coded data corresponding to the historical visit data of each chronic patient in the chronic disease population data set is input into the second cyclic neural network model to obtain the medical visit of the historical visit data of each chronic disease patient in the chronic disease population data set
  • a characterization vector, and the characterization vector of each node in the graph network structure is determined according to the diagnosis characterization vector of the historical diagnosis data of each chronic disease patient.
  • the processor 401 inputs the first coded data corresponding to the historical consultation data of each chronic patient in the chronic disease population data set into the second cyclic neural network model to obtain each chronic disease population data set in the chronic disease population data set.
  • the diagnosis representation vector it is specifically used for:
  • the first coded data corresponding to the historical visit data of each chronic patient in the chronic disease population data set is input into the second cyclic neural network model to obtain the historical chronic disease of each chronic disease patient in the chronic disease population data set
  • the visit characterization vector of each chronic disease patient in the chronic disease population data set is determined.
  • the processor 401 trains the multi-layer perceptron model according to the probability of each chronic disease patient suffering from a complication, and when obtaining a complication prediction model, it is specifically used for:
  • the processor 401 inputs the historical chronic disease data of each chronic disease patient in the chronic disease population data set and the probability of each chronic disease patient suffering from complications into the first recurrent neural network model to obtain the When the medical insurance cost result of each chronic disease patient, it is specifically used for:
  • the chronic disease characterization vector, the complication characterization vector and the probability of each chronic disease patient suffering from the complications of each chronic disease patient in the chronic disease population data set are input into the first recurrent neural network model to obtain all State the chronic disease medical insurance cost result corresponding to the historical chronic disease data of each chronic disease patient in the chronic disease population data set and the complication medical insurance cost result corresponding to the probability of each chronic disease patient suffering from complications;
  • the medical insurance cost result of each chronic disease patient is determined according to the chronic disease medical insurance cost result of each chronic disease patient and the complication medical insurance cost result of each chronic disease patient.
  • the processor 401 trains the first recurrent neural network model to obtain a medical insurance cost prediction model according to the results of the medical insurance cost of each chronic patient, it is specifically used for:
  • the parameters of the first recurrent neural network model are adjusted according to the comparison result, and the chronic disease characterization vector, the complication characterization vector of each chronic disease patient, and the complication characterization vector of each chronic disease patient are adjusted. Probably input the first recurrent neural network model after adjusting the parameters for training, and obtain the medical insurance cost prediction model.
  • the processor 401 inputs the diagnosis data of the chronic disease patient to be tested into the complication prediction model and the medical insurance cost prediction model, and obtains the risk probability of the chronic disease patient to be tested and the medical insurance cost prediction result , Specifically used for:
  • the risk probability of the chronic disease patient to be tested and the chronic disease characterization vector and the complication characterization vector of the chronic disease patient to be tested are input into the medical insurance cost prediction model to obtain the chronic disease patient's chronic disease
  • the processor 401 may be a central processing unit (CenSral Processing UniS, CPU), and the processor may also be other general-purpose processors or digital signal processors (DigiSal Signal Processor, DSP). , Application-specific integrated circuits (ApplicaSion Specific InSegraSed Circuits, ASIC), ready-made programmable gate arrays (Field-Programmable GaSe Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc.
  • the general-purpose processor may be a microprocessor or the processor may also be any conventional processor or the like.
  • the memory 402 may include a read-only memory and a random access memory, and provides instructions and data to the processor 401.
  • a part of the memory 402 may also include a non-volatile random access memory.
  • the memory 402 may also store device type information.
  • the embodiments of the present application also provide a computer-readable storage medium, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the fusion complication described in the embodiment corresponding to FIG. 2 is realized.
  • the risky chronic disease medical insurance cost prediction method can also implement the chronic disease medical insurance cost predicting device for the risk of fusion complications in the embodiment corresponding to FIG. 3 of the present application, which will not be repeated here.
  • the storage medium involved in this application such as a computer-readable storage medium, may be non-volatile or volatile.
  • the computer-readable storage medium may be the internal storage unit of the device for predicting chronic disease medical insurance cost for fusion complications risk described in any of the foregoing embodiments, for example, the hard disk or memory of the device for predicting chronic disease medical insurance cost for fusion complications risk.
  • the computer-readable storage medium may also be an external storage device of the chronic disease medical insurance cost prediction device for the risk of fusion complications, for example, a plug-in hard disk equipped on the chronic disease medical insurance cost prediction device for the risk of fusion complications, Smart memory card (SmarS Media Card, SMC), secure digital (Secure DigiSal, SD) card, flash memory card (Flash Card), etc.
  • the computer-readable storage medium may also include both the internal storage unit of the equipment for predicting the cost of chronic illness medical insurance for the risk of fusion complications and an external storage device.
  • the computer-readable storage medium is used to store the computer program and other programs and data required by the equipment for predicting the cost of chronic illness medical insurance for the risk of fusion complications.
  • the computer-readable storage medium can also be used to temporarily store data that has been output or will be output.
  • the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium.
  • the technical solution of this application is essentially or the part that contributes to the existing technology, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product can be stored in a computer.
  • the read storage medium includes several instructions to enable a computer device (which may be a personal computer, a terminal, 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 media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disks or optical disks and other various programs that can store programs
  • the medium of the code may mainly include a storage program area and a storage data area, where the storage program area may store an operating system, an application program required by at least one function, etc.; the storage data area may store information based on the blockchain node Use the created data, etc.
  • Blockchain essentially a decentralized database, is a series of data blocks associated with cryptographic methods. Each data block contains a batch of network transaction information for verification. The validity of the information (anti-counterfeiting) and the generation of the next block.
  • the blockchain can include the underlying platform of the blockchain, the platform product service layer, and the application service layer.

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Public Health (AREA)
  • Primary Health Care (AREA)
  • Epidemiology (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • General Business, Economics & Management (AREA)
  • Business, Economics & Management (AREA)
  • Data Mining & Analysis (AREA)
  • Pathology (AREA)
  • Databases & Information Systems (AREA)
  • Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
  • Medical Treatment And Welfare Office Work (AREA)

Abstract

L'invention concerne un procédé de prévision de coûts d'assurance maladie associés à une maladie chronique en fonction du risque de complications, et un dispositif associé. Un processeur du dispositif est utilisé pour : déterminer la fréquence de co-occurrence entre différentes complications de chaque patient atteint d'une maladie chronique, et déterminer un vecteur de représentation de chaque noeud dans une structure de réseau de graphe en fonction de la fréquence de co-occurrence ; entrer le vecteur de représentation et des données de maladie chronique historiques dans un modèle de perceptron multicouche, et déterminer un modèle de prévision de complications ; entrer les données de maladie chronique historiques et la probabilité de souffrir de complications pour chaque patient atteint d'une maladie chronique dans un premier modèle de réseau neuronal récurrent, et déterminer un modèle de prévision de coûts d'assurance maladie ; et entrer des données de recours aux soins concernant un patient atteint d'une maladie chronique à tester dans le modèle de prévision de complications et le modèle de prévision de coûts d'assurance maladie, de manière à obtenir la probabilité du risque que le patient atteint d'une maladie chronique à tester souffre de complications et un résultat de prévision de coûts d'assurance maladie. La prévision des coûts d'assurance maladie est réalisée sur la base de la probabilité du risque qu'un patient atteint d'une maladie chronique souffre de complications, ce qui permet d'améliorer la précision de la prévision. Les données peuvent être stockées dans une chaîne de blocs.
PCT/CN2021/083523 2020-11-02 2021-03-29 Procédé de prévision de coûts d'assurance maladie associés à une maladie chronique en fonction du risque de complications, et dispositif associé WO2021203997A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202011206105.4A CN112037918B (zh) 2020-11-02 2020-11-02 一种融合并发症风险的慢病医保费用预测方法及相关设备
CN202011206105.4 2020-11-02

Publications (1)

Publication Number Publication Date
WO2021203997A1 true WO2021203997A1 (fr) 2021-10-14

Family

ID=73573576

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/083523 WO2021203997A1 (fr) 2020-11-02 2021-03-29 Procédé de prévision de coûts d'assurance maladie associés à une maladie chronique en fonction du risque de complications, et dispositif associé

Country Status (2)

Country Link
CN (1) CN112037918B (fr)
WO (1) WO2021203997A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114255074A (zh) * 2021-12-06 2022-03-29 蚂蚁区块链科技(上海)有限公司 基于区块链的评估产品价值的方法及系统

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112037918B (zh) * 2020-11-02 2021-02-12 平安科技（深圳）有限公司 一种融合并发症风险的慢病医保费用预测方法及相关设备

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106407686A (zh) * 2016-09-23 2017-02-15 电子科技大学 一种评估慢性病费用的建模方法
JP6482803B2 (ja) * 2014-09-19 2019-03-13 キヤノンメディカルシステムズ株式会社 シェーマ表示装置
CN109616216A (zh) * 2018-11-30 2019-04-12 平安医疗健康管理股份有限公司 医疗费用预测方法、装置、设备及计算机可读存储介质
CN111275558A (zh) * 2020-01-13 2020-06-12 上海维跃信息科技有限公司 用于确定保险数据的方法和装置
CN111297329A (zh) * 2020-02-24 2020-06-19 苏州大学 预测糖尿病患者心血管并发症动态发病风险的方法及系统
CN112037918A (zh) * 2020-11-02 2020-12-04 平安科技（深圳）有限公司 一种融合并发症风险的慢病医保费用预测方法及相关设备

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104951894B (zh) * 2015-06-25 2018-07-03 成都厚立信息技术有限公司 医院疾病管理智能分析和评估系统
CN107358047A (zh) * 2017-07-13 2017-11-17 刘峰 糖尿病患者评估及管理系统
CN108511070A (zh) * 2018-04-18 2018-09-07 郑州大学第附属医院 一种糖尿病患者评估及管理系统
CN109378074A (zh) * 2018-10-31 2019-02-22 平安医疗健康管理股份有限公司 基于人工智能的慢性病管理方法及相关装置
CN110507296A (zh) * 2019-08-12 2019-11-29 重庆大学 一种基于lstm网络的急性低血压混合预警方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP6482803B2 (ja) * 2014-09-19 2019-03-13 キヤノンメディカルシステムズ株式会社 シェーマ表示装置
CN106407686A (zh) * 2016-09-23 2017-02-15 电子科技大学 一种评估慢性病费用的建模方法
CN109616216A (zh) * 2018-11-30 2019-04-12 平安医疗健康管理股份有限公司 医疗费用预测方法、装置、设备及计算机可读存储介质
CN111275558A (zh) * 2020-01-13 2020-06-12 上海维跃信息科技有限公司 用于确定保险数据的方法和装置
CN111297329A (zh) * 2020-02-24 2020-06-19 苏州大学 预测糖尿病患者心血管并发症动态发病风险的方法及系统
CN112037918A (zh) * 2020-11-02 2020-12-04 平安科技（深圳）有限公司 一种融合并发症风险的慢病医保费用预测方法及相关设备

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114255074A (zh) * 2021-12-06 2022-03-29 蚂蚁区块链科技(上海)有限公司 基于区块链的评估产品价值的方法及系统

Also Published As

Publication number Publication date
CN112037918A (zh) 2020-12-04
CN112037918B (zh) 2021-02-12

Similar Documents

Publication Publication Date Title
Turgeman et al. Insights from a machine learning model for predicting the hospital Length of Stay (LOS) at the time of admission
US20230252017A1 (en) Community data aggregation with automated followup
Taylor et al. Prediction of in‐hospital mortality in emergency department patients with sepsis: a local big data–driven, machine learning approach
Chern et al. Decision tree–based classifier in providing telehealth service
Sim et al. The major effects of health-related quality of life on 5-year survival prediction among lung cancer survivors: applications of machine learning
US11276494B2 (en) Predicting interactions between drugs and diseases
WO2021190661A1 (fr) Système de traitement de données, procédé, appareil et support de stockage
US10318710B2 (en) System and method for identifying healthcare fraud
WO2021203997A1 (fr) Procédé de prévision de coûts d'assurance maladie associés à une maladie chronique en fonction du risque de complications, et dispositif associé
US11398308B2 (en) Physiologic severity of illness score for acute care patients
Calsina-Berna et al. Intrahospital mortality and survival of patients with advanced chronic illnesses in a tertiary hospital identified with the NECPAL CCOMS-ICO© Tool
Golmohammadi et al. Prediction modeling and pattern recognition for patient readmission
Rashidian et al. Deep learning on electronic health records to improve disease coding accuracy
Hu et al. Explainable machine-learning model for prediction of in-hospital mortality in septic patients requiring intensive care unit readmission
Miller et al. Clinicians can independently predict 30-day hospital readmissions as well as the LACE index
US11640857B2 (en) Techniques for providing referrals for opioid use disorder treatment
CN112259180A (zh) 一种基于异构医学知识图谱的疾病预测方法及相关设备
Arnold et al. Can AM-PAC “6-clicks” inpatient functional assessment scores strengthen hospital 30-day readmission prevention strategies?
Luo et al. Identifying frail patients by using electronic health records in primary care: current status and future directions
Shung et al. Neural network predicts need for red blood cell transfusion for patients with acute gastrointestinal bleeding admitted to the intensive care unit
Hilbert et al. Using decision trees to manage hospital readmission risk for acute myocardial infarction, heart failure, and pneumonia
Lai et al. Designing a clinical decision support system to predict readmissions for patients admitted with all-cause conditions
US11848106B1 (en) Clinical event outcome scoring system employing a severity of illness clinical key and method
Oikonomou et al. An explainable machine learning-based phenomapping strategy for adaptive predictive enrichment in randomized clinical trials
Oikonomou et al. An explainable machine learning-based phenomapping strategy for adaptive predictive enrichment in randomized controlled trials

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: 21784648

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: 21784648

Country of ref document: EP

Kind code of ref document: A1