WO2021073263A1 - 一种患病风险的预测方法及装置 - Google Patents

一种患病风险的预测方法及装置 Download PDF

Info

Publication number
WO2021073263A1
WO2021073263A1 PCT/CN2020/111662 CN2020111662W WO2021073263A1 WO 2021073263 A1 WO2021073263 A1 WO 2021073263A1 CN 2020111662 W CN2020111662 W CN 2020111662W WO 2021073263 A1 WO2021073263 A1 WO 2021073263A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
risk
physical examination
key factors
target
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/CN2020/111662
Other languages
English (en)
French (fr)
Inventor
代心灵
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 WO2021073263A1 publication Critical patent/WO2021073263A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

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
    • 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
    • G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
    • 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

  • This application relates to the field of artificial intelligence technology, and in particular to a method and device for predicting disease risk.
  • Cardiovascular disease is the primary disease burden for human health in the world. The morbidity and mortality of cardiovascular diseases in my country are showing a rapid upward trend. It is estimated that hypertension, hyperlipidemia, and hyperglycemia are the three most common cardiovascular diseases. It can also be said that the "three highs” disease is the culprit of most other cardiovascular and cerebrovascular diseases. Therefore, the prevention and treatment of the "three highs” disease has become Preventing cardiovascular diseases and safeguarding the important needs of physical health, the current "three highs" preventive examinations in hospitals are obtained based on the specific data of traditional measurements and the professional knowledge of doctors.
  • an embodiment of the present application provides a method for predicting disease risk, which may include: receiving key factors, which are used to indicate the type of disease of the target disease or the influencing factors that affect the target disease;
  • the key factors perform data screening and processing on the physical examination data of the target user to obtain risk data.
  • the risk data includes one or more of blood pressure data, blood lipid data, and blood glucose data; according to the risk data, a target scoring model is adopted , Determining the risk score of the target user, the target scoring model is a prediction model trained based on the Xgboost algorithm and the K-Folds cross-validation algorithm method; and determining the risk level of the target user's disease risk according to the risk score.
  • an embodiment of the present application provides a device for predicting disease risk, which may include: a receiving unit, configured to receive key factors, the key factors being used to indicate the type of disease of the target disease or to indicate that the target disease is affected
  • the processing unit is used to screen and process the physical examination data of the target user according to the key factors to obtain risk data.
  • the risk data includes one or more of blood pressure data, blood lipid data, and blood glucose data; according to the risk data, pass The target scoring model determines the risk score of the target user, the target scoring model is a prediction model trained based on the Xgboost algorithm and the K-Folds cross-validation algorithm method; the risk of the target user is determined according to the risk score Risk level.
  • 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 steps: receiving key factors, the key The factor is used to indicate the disease type of the target disease or the influencing factor that affects the target disease; the physical examination data of the target user is screened and processed according to the key factors to obtain risk data, the risk data includes blood pressure data, blood lipids One or more of data and blood glucose data; according to the risk data, the risk score of the target user is determined through a target scoring model, the target scoring model is trained based on the Xgboost algorithm and the K-Folds cross-validation algorithm method A prediction model; the risk level of the target user's disease risk is determined according to the risk score.
  • Using the embodiments of the present application can effectively predict the risk of a user suffering from diseases, and further, can timely and effectively identify the potential threats of chronic diseases caused by the "three highs" factors in the body.
  • FIG. 2A is a schematic diagram of the flow of a method for predicting disease risk according to an embodiment of the present application.
  • Fig. 2B is a schematic diagram of an application interface of a terminal device provided by an embodiment of the present application.
  • Fig. 3 is a schematic diagram of the flow of a method for predicting disease risk provided by an embodiment of the present application.
  • Fig. 4 is a schematic structural diagram of a device for predicting disease risk provided by an embodiment of the present application.
  • the technical solution of the present application can be applied to the fields of artificial intelligence, smart city and/or digital medical technology, and can predict the risk of a user's illness, so as to perform health management and realize smart medical treatment.
  • server used in this application are used to denote computer-related entities, hardware, firmware, a combination of hardware and software, software, or software in execution.
  • the server may be, but is not limited to, a processor, a data processing platform, a computing device, a computer, two or more computers, etc.
  • XGBoost XtremeGradient Boosting
  • CART regression tree
  • the regression tree refers to the construction of a classification tree based on training features and training data to determine the prediction result of each piece of data. Among them, the construction of the tree requires the use of Gini index (Gini index) to calculate the gain, that is, the feature selection of the construction tree.
  • Gini index Gini index
  • “Three high” diseases refer to hypertension, hyperglycemia (diabetes) and hyperlipidemia. They are "diseases of wealth” derived from modern society, and they may exist alone, or they may be related to each other. For example, people with diabetes are prone to suffer from high blood pressure or hyperlipidemia at the same time, and hyperlipidemia is the main factor in the formation and development of arteriosclerosis. In patients with arteriosclerosis, poor blood vessel elasticity increases blood pressure. Therefore, if any one of these three diseases appears, the "three highs" disease will easily develop in the later stage.
  • the device 101 for predicting disease risk may include, but is not limited to, a background server, a component server, a data processing server, etc.
  • the server can communicate with multiple terminals via the Internet and receive The medical examination data from the terminal equipment provides the service of disease risk prediction for terminal users.
  • the server may receive key factors, screen and process the physical examination data of the target user according to the key factors, and obtain risk data.
  • the risk data includes one or more of blood pressure data, blood lipid data, and blood glucose data;
  • the risk data determines the risk score of the target user through a target scoring model, the target scoring model is a prediction model trained based on the Xgboost algorithm and the K-Folds cross-validation algorithm method; the target user is determined according to the risk score The risk level of the risk of illness.
  • the terminal device 102 can install and run related applications.
  • Application refers to a program that corresponds to a server and provides local services to customers.
  • the local service may include, but is not limited to: sending physical examination data to the server and receiving a prediction report sent by the server (for example, the prediction report may include the user's risk score and risk level), other shared information, and so on.
  • the terminal in the embodiment of this solution can include, but is not limited to, any electronic product based on a smart operating system, which can interact with the user through input devices such as keyboard, virtual keyboard, touch pad, touch screen, and voice control device, such as Smartphones, tablets, personal computers, etc.
  • smart operating systems include, but are not limited to, any operating system that enriches device functions by providing various mobile applications to mobile devices, such as AndroidTM, iOSTM, Windows PhoneTM, etc.
  • FIG. 2A is a schematic diagram of a method for predicting disease risk according to an embodiment of the present application. It can be applied to the system in FIG. 1 described above, and the following description will be made from one side of the device 101 for predicting the risk of illness in conjunction with FIG. 2A.
  • the method may include the following steps S201-S203.
  • Step S201 receiving the key factors, and performing data screening and processing on the medical examination data of the target user according to the key factors to obtain risk data.
  • the device 101 for predicting disease risk may receive key factors, and perform data screening and processing on the physical examination data of the target user according to the key factors to obtain risk data, wherein the key factors are used to indicate the disease type of the target disease Or indicate an influencing factor that affects the target disease.
  • the key factor may include a keyword or a key factor
  • the risk data includes one or more of blood pressure data, blood lipid data, and blood glucose data.
  • the key factors include the influencing factors that affect the target disease.
  • the key factors can be the influencing factors that determine certain risk data that affect the target disease in the physical examination data, or the influencing factors that affect the target disease in the physical examination data. The influencing factors of the target disease are mentioned.
  • the physical examination data can include basic surgical examination data, basic internal medicine examination data, oral examination data, ultrasound data, electrocardiogram data, ophthalmological examination data, gynecological examination data, biochemical examination data, blood pressure measurement results, blood lipid measurement results, and blood glucose measurement One or more of the results.
  • Step S202 Determine the risk score of the target user through the target scoring model according to the risk data.
  • the device 101 for predicting disease risk may determine the risk score corresponding to the target user through a target scoring model based on the disease risk data, where the target scoring model is based on the Xgboost algorithm and K-Folds cross-validation The prediction model trained by the algorithm method.
  • the key factor includes M keywords, and the M keywords are used to indicate the disease type of the target disease, where M is a positive integer greater than or equal to 1; therefore, M keywords are received, Perform data screening on the physical examination data according to the M keywords, and determine that the physical examination data corresponding to the M keywords that are screened out are the risk data.
  • M is a positive integer greater than or equal to 1
  • relevant feature data such as ophthalmology and oral cavity in the physical examination data can be deleted, while the physical examination data about blood pressure measurement results, blood lipid measurement results, and blood glucose measurement results can be retained.
  • the key factor includes M keywords, and the M keywords are used to indicate the disease type of the target disease, where M is a positive integer greater than or equal to 1; therefore, M keywords are received,
  • the physical examination data corresponding to the M keywords are processed into 0-1 characteristic data according to a regular expression manner, and the risk data is obtained.
  • the keyword is blood pressure, blood sugar or blood lipids, etc.
  • some words like "fatty liver” and "hyperglycemia” form a 0-1 feature, that is, high blood sugar is considered as 1, which is the value of blood glucose measurement.
  • the type feature can remain unchanged; or, the text "blood pressure measurement result is normal” in the physical examination data can also be written to form a 0-1 feature, that is, the blood pressure measurement result is normal to 0.
  • Step S203 Determine the risk level of the target user's disease risk according to the risk score.
  • the device 101 for predicting disease risk may determine the risk level of the target user's disease risk according to the risk score.
  • the device 101 for predicting disease risk determines a disease risk prediction report according to the risk score, and the disease risk prediction report includes the risk level of the target user's disease risk corresponding to the risk score.
  • FIG. 2B is a schematic diagram of an application interface of a terminal device according to an embodiment of the present application.
  • the target user can use the terminal device 102 to predict the disease risk of the acquired physical examination report, so as to obtain the user's disease risk prediction report, and timely understand the level of the disease that he or she may suffer from, so as to seek medical treatment early for prevention.
  • the disease risk prediction report may also include preventive measures or treatment methods required by the user corresponding to the risk level at this stage.
  • the risk score can be determined based on the target scoring model based on the user's physical examination data, and then the disease risk level can be determined according to the risk score, and finally a disease risk prediction report can be generated.
  • the preventive measures or treatment required for each user's stage can be obtained Means, especially for some potential patients, targeted prevention and intervention based on some potential threat factors found by the above algorithm.
  • Implementing the embodiments of this application can receive key factors, and perform data screening and processing on the physical examination data of the target user according to the key factors to obtain one or more of blood pressure data, blood lipid data, and blood glucose data; according to the processed data Determine the risk score corresponding to the target user through a target scoring model trained based on the Xgboost algorithm and the K-Folds cross-validation algorithm method; determine the risk level of the target user's disease risk according to the risk score.
  • the risk prediction model can effectively predict the user’s risk of illness, and furthermore, it can be timely and effectively identified
  • chronic diseases such as high blood pressure, hyperglycemia, hyperlipidemia, cardiovascular and cerebrovascular diseases, etc.
  • Step S301 Obtain physical examination data of N users as sample physical examination data.
  • the device 101 for predicting the risk of illness may obtain the physical examination data of N users from the physical examination center or other data platforms as sample physical examination data. Because the physical examination data of N users is the sample data of the training target scoring model, the The physical examination data of N users does not include the physical examination data of the target user, and N is a positive integer greater than 1.
  • Step S302 According to the K-Folds cross-validation algorithm, the obtained N sample physical examination data are divided into K-1 training sets and 1 test set.
  • the device 101 for predicting disease risk divides the acquired physical examination data of N samples into K-1 training sets and 1 test set according to the K-Folds cross-validation algorithm, where K is a positive integer greater than 1.
  • K-Folds cross-validation algorithm is used to obtain K training sets and test sets from N sample physical examination data: the data set is divided into K parts, so that K-1 parts of the data set are used to train the model each time, and the remaining 1 part of the data is used to verify the model effect .
  • Step S303 Based on the Xgboost algorithm, establish i regression tree training prediction models on the K-1 training sets.
  • the objective function in the training process can be optimized: the predicted value of the tree group is as close as possible to the true value (ie: accuracy) and as much as possible Great generalization ability.
  • the prediction error in the objective function should be as small as possible, that is, the leaf nodes of the tree should be as few as possible, and the node values should not be extremely distributed as much as possible.
  • the eigenvalue that minimizes its loss function is found to be split step by step, and the split node of each eigenvalue is determined by the node that minimizes the root mean square error of the eigenvalue.
  • Step S304 According to a test set test, the risk of illness is scored and predicted through the trained prediction model.
  • the device 101 for predicting the risk of disease performs scoring prediction on the size of the risk of disease according to a test set test and a trained prediction model. That is, after the prediction model is trained from the K-1 training set, the remaining 1 test set data needs to be substituted into the trained prediction model to predict the risk of illness for score prediction. If the result of the score prediction is not within the preset error range, Therefore, the prediction model cannot accurately predict the user's risk of illness, and the prediction model needs further training.
  • Step S305 If the result of the scoring prediction is within the preset error range, the prediction model is determined to be the target scoring model.
  • the device 101 for predicting disease risk substitutes the remaining 1 test set data into the trained prediction model to perform scoring prediction of the risk of disease, if the result of the score prediction is within the preset error range, it is determined
  • the prediction model is the target scoring model, which proves that the prediction model can accurately predict the user's risk of illness.
  • the device 101 for predicting the risk of illness may also determine the risk score corresponding to the N sample physical examination data according to the N sample physical examination data and the target scoring model, and determine the N sample physical examination data according to the risk score Corresponding to the risk score, divide the risk level and determine the mapping relationship between the risk score and the risk level. It is understandable that the higher the risk score, the higher the risk level corresponding to the device, and the greater the probability that the user may develop a disease. Therefore, according to the mapping relationship, the risk level corresponding to the risk score of the target user can be determined, and the probability of the target user suffering from the disease can be determined.
  • the device 101 for predicting disease risk is based on the Xgboost algorithm, establishes i regression trees on the K training sets, and determines s different key factors in the physical examination data that affect the risk data, where s Is a positive integer greater than 1.
  • the key factors include the s different key factors;
  • the data screening and processing of the physical examination data of the target user according to the key factors to obtain risk data includes: determining the influence of the s key factors
  • Data screening is performed on the physical examination data to obtain the risk data.
  • the physical examination data of the target user corresponding to the first j key factors can be filtered out of the physical examination data of the target user, and then the final risk data can be obtained after data processing.
  • the risk data can be expressed in a regular manner. Processed into 0-1 feature data or converted text data into numerical data.
  • Step S308 Determine the risk level of the target user's disease risk according to the risk score.
  • step S307 to step S308 reference may also be made to the related description of step S202 to step S203 in FIG. 2A, which will not be repeated here.
  • the following provides a device for predicting disease risk related to the embodiment of the application.
  • the device 10 for predicting disease risk may be a method of obtaining, processing, analyzing and extracting valuable Data, which brings all kinds of convenient service equipment for third-party use.
  • FIG. 4 is a schematic structural diagram of a device for predicting disease risk according to an embodiment of the present application. It may include a receiving unit 401, a processing unit 402, a first determining unit 403, and a second determining unit 404, and may also include a model training unit 405, a third determining unit 406, and a fourth determining unit 407.
  • the processing unit 402 is configured to perform data screening and processing on the physical examination data of the target user according to the key factors to obtain risk data.
  • the risk data includes one or more of blood pressure data, blood lipid data, and blood glucose data.
  • the first determining unit 403 is configured to determine the risk score of the target user through a target scoring model according to the risk data, and the target scoring model is a prediction model trained based on the Xgboost algorithm and the K-Folds cross-validation algorithm device.
  • the second determining unit 404 is configured to determine the risk level of the target user's disease risk according to the risk score.
  • the physical examination data includes basic surgical examination data, basic internal medicine examination data, oral examination data, ultrasound data, electrocardiogram data, ophthalmological examination data, gynecological examination data, biochemical examination data, blood pressure measurement results, One or more of blood lipid measurement results and blood glucose measurement results.
  • the key factors include M keywords, and the M keywords are used to indicate the disease type of the target disease, where M is a positive integer greater than or equal to 1;
  • the processing unit 402 is specifically configured to: process the physical examination data corresponding to the M keywords into 0-1 characteristic data according to a regular expression manner, and obtain the risk data.
  • the device further includes: a third determining unit 406, configured to establish i regression trees on the K training sets based on the Xgboost algorithm, and determine that the physical examination data affects the S different key factors of the risk data, where s is a positive integer greater than 1; the key factors include the s different key factors; the processing unit 402 is specifically configured to: determine the s key factors The physical examination data of the target user corresponding to the first j key factors with the largest influence value in the influence value of the factor, where j is a positive integer greater than 1 and less than s; corresponding to the first j key factors Data screening is performed on the physical examination data of the target user to obtain the risk data.
  • a third determining unit 406 configured to establish i regression trees on the K training sets based on the Xgboost algorithm, and determine that the physical examination data affects the S different key factors of the risk data, where s is a positive integer greater than 1; the key factors include the s different key factors; the processing unit 402 is specifically configured
  • the device further includes: a fourth determining unit 407, configured to determine the risk score corresponding to the N sample physical examination data according to the N sample physical examination data and the target scoring model ; According to the determination of the risk scores corresponding to the N sample physical examination data, the risk levels are divided and the mapping relationship between the risk scores and the risk levels is determined.
  • a fourth determining unit 407 configured to determine the risk score corresponding to the N sample physical examination data according to the N sample physical examination data and the target scoring model ; According to the determination of the risk scores corresponding to the N sample physical examination data, the risk levels are divided and the mapping relationship between the risk scores and the risk levels is determined.
  • each operation can also correspond to the corresponding description of the method embodiment shown in FIG. 2A to FIG. 3, which will not be repeated here.
  • FIG. 5 is a schematic structural diagram of another apparatus for predicting disease risk according to an embodiment of the present application.
  • the apparatus 20 includes at least one processor 501, at least one memory 502, and at least one communication interface 503.
  • the device may also include general components such as antennas, which will not be described in detail here.
  • the processor 501 may be a general-purpose central processing unit (CPU), a microprocessor, or an application-specific integrated circuit (application-specific integrated circuit). integrated circuit, ASIC), or one or more integrated circuits used to control the execution of the above program programs.
  • CPU central processing unit
  • microprocessor or an application-specific integrated circuit (application-specific integrated circuit). integrated circuit, ASIC), or one or more integrated circuits used to control the execution of the above program programs.
  • ASIC application-specific integrated circuit
  • the memory 502 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, random access memory (RAM), or other types that can store information and instructions
  • the dynamic storage device can also be an electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, EEPROM), a compact disc (Compact Disc Read-Only Memory, CD-ROM) or other optical disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or can be used to carry or Any other medium that stores desired program codes in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.
  • the memory can exist independently and is connected to the processor through a bus.
  • the memory can also be integrated with the processor.
  • each functional unit in the device 20 for predicting the risk of illness described in the embodiment of the present application can refer to the corresponding description of the method embodiment shown in FIGS. 2A to 3, which will not be repeated here.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed in multiple locations. On a network unit. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments of the present application.
  • the functional components in the various embodiments of the present application may be integrated into one component, or each component may exist alone physically, or two or more components may be integrated into one component.
  • the above-mentioned integrated components can be implemented in the form of hardware or software functional units.
  • the integrated component 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 is stored in a storage medium. It includes several instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in the embodiments of the present application.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disks or optical disks and other media that can store program codes.

Landscapes

  • Engineering & Computer Science (AREA)
  • Medical Informatics (AREA)
  • Public Health (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Epidemiology (AREA)
  • Data Mining & Analysis (AREA)
  • Biomedical Technology (AREA)
  • Databases & Information Systems (AREA)
  • Pathology (AREA)
  • Medical Treatment And Welfare Office Work (AREA)
  • Measuring And Recording Apparatus For Diagnosis (AREA)

Abstract

一种患病风险的预测方法及装置，其中，一种患病风险的预测方法，可包括：接收关键因素，根据所述关键因素对目标用户的体检数据进行数据筛选并处理，获得风险数据，所述风险数据包括血压数据、血脂数据以及血糖数据中的一种或多种；根据所述风险数据通过目标评分模型，确定所述目标用户的风险分数，所述目标评分模型为基于Xgboost算法和K-Folds交叉验证算法方法训练出的预测模型；根据所述风险分数确定所述目标用户患病风险的风险等级。该方法可以有效的预测该用户患病的风险大小，进一步的，可以及时有效的识别身体内造成疾病的潜在威胁，尽早对疾病进行早期筛查和干预，以便延缓疾病显性化和防止慢病重症化。

Description

一种患病风险的预测方法及装置
本申请要求于2019年10月15日提交中国专利局、申请号为201910980973.9，发明名称为“一种患病风险的预测方法及装置”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
本申请涉及人工智能技术领域，尤其涉及一种患病风险的预测方法及装置。
 背景技术
心血管疾病是全球人类健康的首要疾病负担。我国心血管病的发病率和死亡率呈快速上升趋势。据估计高血压、高血脂、高血糖这最常见的三种心血管疾病，也可以说“三高”疾病是其他大多数心脑血管疾病的元凶，因此，“三高”疾病的防治已经成为预防心血管疾病、保障身体健康的重要需求，目前医院的“三高”预防检查是根据传统测量的具体数据和医生的专业知识来获得。但是发明人意识到，对于“三高”疾病患者，仅某项影响因素指标超标时，“三高”疾病的特性有可能未能及时展现出来，导致人们对于患病的风险认知并不清楚，不能及时防治。
因此根据传统测量的具体数据和医生的专业知识来诊断防治疾病的方法就显得为时已晚。进一步的，一旦患病后再治疗，对于用户的身体及经济成本都会造成很大的损失。所以，如何及时有效的识别身体内造成疾病的潜在威胁，尽早对疾病产生的原因进行早期筛查和干预，以便延缓疾病显性化和防止慢病重症化。
 技术问题
鉴于上述问题，提出了本申请以便提供一种克服上述问题或者至少部分地解决上述问题的一种患病风险的预测方法及装置。
 技术解决方案
第一方面，本申请实施例提供了一种患病风险的预测方法，可包括：接收关键因素，所述关键因素用于指示目标疾病的疾病类型或指示影响所述目标疾病的影响因子；根据所述关键因素对目标用户的体检数据进行数据筛选并处理，获得风险数据，所述风险数据包括血压数据、血脂数据以及血糖数据中的一种或多种；根据所述风险数据通过目标评分模型，确定所述目标用户的风险分数，所述目标评分模型为基于Xgboost算法和K-Folds交叉验证算法方法训练出的预测模型；根据所述风险分数确定所述目标用户患病风险的风险等级。
第二方面，本申请实施例提供了一种患病风险的预测装置，可包括：接收单元，用于接收关键因素，所述关键因素用于指示目标疾病的疾病类型或指示影响所述目标疾病的影响因子；处理单元，用于根据所述关键因素对目标用户的体检数据进行数据筛选并处理，获得风险数据，所述风险数据包括血压数据、血脂数据以及血糖数据中的一种或多种；第一确定单元，用于根据所述风险数据通过目标评分模型，确定所述目标用户的风险分数，所述目标评分模型为基于Xgboost算法和K-Folds交叉验证算法装置训练出的预测模型；第二确定单元，用于根据所述风险分数确定所述目标用户患病风险的风险等级。
第三方面，本申请实施例提供了一种患病风险的预测装置，包括存储组件，处理组件和通信组件，存储组件，处理组件和通信组件相互连接，其中，存储组件用于存储计算机程序，通信组件用于与外部设备进行信息交互；处理组件被配置用于调用计算机程序，执行以下步骤：接收关键因素，所述关键因素用于指示目标疾病的疾病类型或指示影响所述目标疾病的影响因子；根据所述关键因素对目标用户的体检数据进行数据筛选并处理，获得风险数据，所述风险数据包括血压数据、血脂数据以及血糖数据中的一种或多种；根据所述风险数据通过目标评分模型，确定所述目标用户的风险分数，所述目标评分模型为基于Xgboost算法和K-Folds交叉验证算法方法训练出的预测模型；根据所述风险分数确定所述目标用户患病风险的风险等级。
第四方面，本申请实施例提供了一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，所述计算机程序被处理器执行以实现以下步骤：接收关键因素，所述关键因素用于指示目标疾病的疾病类型或指示影响所述目标疾病的影响因子；根据所述关键因素对目标用户的体检数据进行数据筛选并处理，获得风险数据，所述风险数据包括血压数据、血脂数据以及血糖数据中的一种或多种；根据所述风险数据通过目标评分模型，确定所述目标用户的风险分数，所述目标评分模型为基于Xgboost算法和K-Folds交叉验证算法方法训练出的预测模型；根据所述风险分数确定所述目标用户患病风险的风险等级。
 有益效果
采用本申请实施例可以有效的预测用户患病的风险大小，进一步的，可以及时有效的识别身体内因“三高”因素等造成慢性疾病的潜在威胁。
 附图说明
为了更清楚地说明本申请实施例或背景技术中的技术方案，下面将对本申请实施例或背景技术中所需要使用的附图进行说明。
图1是本申请实施例提供的一种患病风险的预测系统架构的示意图。
图2A是本申请实施例提供的一种患病风险的预测方法流程的示意图。
图2B是本申请实施例提供的一种终端设备的应用界面示意图。
图3是本申请实施例提供的一种患病风险的预测方法流程的示意图。
图4是本申请实施例提供的一种患病风险的预测装置的结构示意图。
图5是本申请实施例提供的另一种患病风险的预测装置的结构示意图。
 本发明的实施方式
下面将结合本申请实施例中的附图，对本申请实施例进行描述。
本申请的技术方案可应用于人工智能、智慧城市和/或数字医疗技术领域，可预测用户患病的风险，以进行健康管理，实现智慧医疗。
本申请的说明书和权利要求书及所述附图中的术语“第一”、“第二”和“第三”等是用于区别不同对象，而不是用于描述特定顺序。此外，“包括”和“具有”以及它们任何变形，意图在于覆盖不排他的包含。例如包含了一系列步骤或单元的过程、方法、系统、产品或设备没有限定于已列出的步骤或单元，而是可选地还包括没有列出的步骤或单元，或可选地还包括对于这些过程、方法、产品或设备固有的其它步骤或单元。
在本文中提及“实施例”意味着，结合实施例描述的特定特征、结构或特性可以包含在本申请的至少一个实施例中。在说明书中的各个位置出现该短语并不一定均是指相同的实施例，也不是与其它实施例互斥的独立的或备选的实施例。本领域技术人员显式地和隐式地理解的是，本文所描述的实施例可以与其它实施例相结合。
在本申请中使用的术语“服务器”、“单元”、“系统”等用于表示计算机相关的实体、硬件、固件、硬件和软件的组合、软件、或执行中的软件。例如，服务器可以是但不限于，处理器，数据处理平台，计算设备，计算机，两个或更多个计算机等。
首先，对本申请中的部分用语进行解释说明，以便于本领域技术人员理解。
（1）XGBoost算法，全称是XtremeGradient Boosting，其中回归树(regressiontree，CART)是XGBoost最基本的组成部分，回归树是指其根据训练特征及训练数据构建分类树，判定每条数据的预测结果。其中，构建树需要使用吉尼系数（Gini index）计算增益，即进行构建树的特征选取。
（2）K-Folds交叉验证算法，又称循环验证。它将原始数据分成K组(K-Fold)，将每个子集数据分别做一次验证集，其余的K-1组子集数据作为训练集，这样会得到K个模型。这K个模型分别在验证集中评估结果，最后的误差MSE(Mean Squared Error)加和平均就得到交叉验证误差。交叉验证有效利用了有限的数据，并且评估结果能够尽可能接近模型在测试集上的表现，可以做为模型优化的指标使用。
（3）“三高”疾病，是指高血压、高血糖（糖尿病）和高脂血症。它们是现代社会所派生出来的“富贵病”，可能单独存在，也可能相互关联。如：糖尿病人很容易同时患上高血压或高血脂症，而高血脂又是动脉硬化形成和发展的主要因素，动脉硬化患者血管弹性差加剧血压升高。所以，出现这三种疾患中的任何一种，后期都易形成了“三高”疾病。
其次，对本申请实施例所基于的其中一种患病风险的预测系统架构进行描述。请参考附图1，图1是本申请实施例提供的一种患病风险的预测系统架构的示意图，包括：患病风险的预测装置101和终端装置102。
患病风险的预测装置101可以包括但不限于后台服务器、组件服务器、数据处理服务器等，当上述患病风险的预测装置101为服务器时，所述服务器可以通过互联网与多个终端进行通信，接收来自终端设备的体检数据，为终端的用户提供患病风险预测的服务。例如，服务器可以接收关键因素，根据关键因素对目标用户的体检数据进行数据筛选并处理，获得风险数据，所述风险数据包括血压数据、血脂数据以及血糖数据中的一种或多种；根据所述风险数据通过目标评分模型，确定所述目标用户的风险分数，所述目标评分模型为基于Xgboost算法和K-Folds交叉验证算法方法训练出的预测模型；根据所述风险分数确定所述目标用户患病风险的风险等级。
终端设备102可以安装并运行相关的应用。应用是指与服务器相对应，为客户提供本地服务的程序。这里，该本地服务可包括但不限于：向服务器发送体检数据和接收服务器发送的预测报告（例如：预测报告中可以包含用户的风险分数和风险等级）以及其他共享信息等等。本方案实施例中的终端可以包括但不限于任何一种基于智能操作系统的电子产品，其可与用户通过键盘、虚拟键盘、触摸板、触摸屏以及声控设备等输入设备来进行人机交互，诸如智能手机、平板电脑、个人电脑等。其中，智能操作系统包括但不限于任何通过向移动设备提供各种移动应用来丰富设备功能的操作系统，诸如：安卓 (Android™)、iOS™、Windows Phone™等。
还可以理解的是，图1的患病风险的预测系统架构只是本申请实施例中的部分示例性的实施方式，本申请实施例中的患病风险的预测系统架构包括但不仅限于以上患病风险的预测系统架构。
参考附图2A，图2A是本申请实施例提供的一种患病风险的预测方法流程的示意图。可应用于上述图1中的系统，下面将结合图2A从患病风险的预测装置101的单侧进行描述。该方法可以包括以下步骤S201-步骤S203。
步骤S201：接收关键因素，根据关键因素对目标用户的体检数据进行数据筛选并处理，获得风险数据。
具体地，患病风险的预测装置101可以接收关键因素，根据所述关键因素对目标用户的体检数据进行数据筛选并处理，获得风险数据，其中，所述关键因素用于指示目标疾病的疾病类型或指示影响所述目标疾病的影响因子，例如：所述关键因素可以包括关键字或者关键因子，所述风险数据包括血压数据、血脂数据以及血糖数据中的一种或多种。可以理解的是，所述关键因素包括的影响所述目标疾病的影响因子是关键因子是可以是在体检数据中确定某些影响目标疾病的风险数据的影响因子还可以是在体检数据中影响所述目标疾病的影响因子，因此关键因素主要是指示能够对目标疾病造成影响的因素，或者是指示目标疾病的疾病类型，进一步的，可以通过该疾病类型可以筛选出目标疾病造成影响的因素。还可以理解的是，因为体检数据多种多样（例如：文字形式、图片形式、曲线形式等），所以不能够直接输入目标评分模型中；又因为体检数据往往是全面的身体检查数据，如果需要预估某一类型的疾病需要把与该疾病类型相关的体检数据筛选出来后，在针对筛选出的体检数据做患病风险预测，由此得到的风险分数才会更准确。因此，本申请实施需要对所述体检数据进行数据处理，再将处理后的数据输入目标评分模型中。
可选的，体检数据可以包括外科基本检查数据、内科基本检查数据、口腔检查数据、超声数据、心电图数据、眼科检查数据、妇科检查数据、生化检查数据、血压测量结果、血脂测量结果以及血糖测量结果中的一种或多种。
步骤S202：根据风险数据通过目标评分模型，确定目标用户的风险分数。
具体地，患病风险的预测装置101可以根据所述患病风险数据通过目标评分模型，确定所述目标用户对应的风险分数，其中，所述目标评分模型为基于Xgboost算法和K-Folds交叉验证算法方法训练出的预测模型。
可选的，所述关键因素包括M个关键字，所述M个关键字用于指示所述目标疾病的疾病类型，其中，M为大于或等于1的正整数；因此接收M个关键字，根据所述M个关键字对所述体检数据进行数据筛选，确定筛选出的与所述M个关键字对应的体检数据为所述风险数据。例如：关键字是血压、血糖或血脂等，则体检数据中如眼科、口腔等相关特征数据可以删除，而关于血压测量结果、血脂测量结果以及血糖测量结果的体检数据可以保留。
可选的，所述关键因素包括M个关键字，所述M个关键字用于指示所述目标疾病的疾病类型，其中，M为大于或等于1的正整数；因此接收M个关键字，将所述M个关键字对应的所述体检数据按照正则表达方式处理为0-1特征数据，获得所述风险数据。例如：关键字是血压、血糖或血脂等，则体检数据中，一些像＂脂肪肝＂，＂高血糖＂这样的文字写正则构成0-1特征，即高血糖认为1，其中血糖测量的数值型特征可以保持不变；又或者，体检数据中“血压测量结果正常”的文字也可以写正则构成0-1特征，即血压测量结果正常为0。
步骤S203：根据风险分数确定目标用户患病风险的风险等级。
具体地，患病风险的预测装置101可以根据所述风险分数确定所述目标用户患病风险的风险等级。
可选的，患病风险的预测装置101根据所述风险分数确定患病风险预测报告，所述患病风险预测报告包括所述风险分数对应的所述目标用户患病风险的风险等级。请参考附图2B，图2B是本申请实施例提供的一种终端设备的应用界面示意图。目标用户可以通过终端设备102将获取的体检报告进行患病风险预测，以获取该用户的患病风险预测报告，及时了解自身的可能会患的疾病的等级，以便提早就医预防。其中，所述患病风险预测报告还可以包含所述风险等级对应的用户在该阶段需要的预防措施或治疗手段。例如：可以根据用户体检数据基于目标评分模型确定风险分数后，再根据风险分数确定患病风险等级，最后生成患病风险预测报告，根据该报告得到每个用户所处阶段需要的预防措施或治疗手段，尤其对于一些潜在患者，根据以上算法找到的一些潜在威胁因子有针对性地进行预防及干预。
实施本申请实施例，可以接收关键因素，根据所述关键因素对目标用户的体检数据进行数据筛选并处理，获得血压数据、血脂数据以及血糖数据中的一种或多种；根据处理好的数据，通过基于Xgboost算法和K-Folds交叉验证算法方法训练出的目标评分模型，确定所述目标用户对应的风险分数；根据所述风险分数确定所述目标用户患病风险的风险等级。将处理好的血压数据、血脂数据以及血糖数据中的一种或多种输入到风险预测模型中后，风险预测模型可以有效的预测该用户患病的风险大小，进一步的，可以及时有效的识别身体内因“三高”因素等造成慢性疾病（例如：高血压、高血糖、高血脂、心脑血管疾病等）的潜在威胁，尽早对造成该类型疾病的因素进行早期筛查和干预，以便延缓疾病显性化和防止慢病重症化。
参考附图3，图3是本申请实施例提供的一种患病风险的预测方法流程的示意图。可应用于上述图1中的系统，下面将结合图3从患病风险的预测装置101的单侧进行描述。该方法可以包括以下步骤S301-步骤S308。
步骤S301：获取N个用户的体检数据为样本体检数据。
具体地，患病风险的预测装置101可以从体检中心或者其他数据平台获取N个用户的体检数据为样本体检数据，因为N个用户的体检数据是作为训练目标评分模型的样本数据，所以所述N个用户的体检数据不包括目标用户的体检数据，N为大于1的正整数。
步骤S302：根据K-Folds交叉验证算法将获取的N个样本体检数据分成K-1个训练集及1个测试集。
具体地，患病风险的预测装置101根据所述K-Folds交叉验证算法将获取的N个样本体检数据分成K-1个训练集及1个测试集，其中，K为大于1的正整数。通过K-Folds交叉验证算法在N个样本体检数据中得到K个训练集及测试集：将数据集分成K份，以满足每次用其中K-1份训练模型，剩余1份数据验证模型效果。
步骤S303：基于Xgboost算法，在所述K-1个训练集上建立i个回归树训练预测模型。
具体地，患病风险的预测装置101基于Xgboost算法，在所述K-1个训练集上建立i个回归树训练预测模型，所述预测模型用于对患病的风险大小进行评分预测，其中，i为大于1的正整数。即，在训练集上建立m个回归树进行预测：每一个回归树都是一个弱学习器，通过一步步加入新的弱学习器，来努力纠正前面所有弱学习器的残差，因此，最终这m个学习器相加在一起可以用来进行最终患病风险的预测。
可选的，在患病风险的预测装置101基于Xgboost算法训练模型的过程中，可以优化该训练过程中的目标函数：使得树群的预测值尽量接近真实值（即：准确率）而且有尽量大的泛化能力。需要说明的是，要满足目标函数中预测误差尽量小，即树的叶子节点尽量少，节点数值尽量不极端分布。Xgboost算法的优化过程中，找到使其损失函数最小的特征值一步步进行分裂，每个特征值的分裂节点由使得特征值的均方根误差最小的节点确定。
步骤S304：根据1个测试集测试，通过训练后的预测模型对患病的风险大小进行评分预测。
具体地，患病风险的预测装置101根据1个测试集测试，通过训练后的预测模型对患病的风险大小进行评分预测。即K-1个训练集训练出预测模型后需要将剩余的1个测试集数据代入训练好的预测模型中进行患病的风险大小进行评分预测，若评分预测的结果不在预设误差范围内，则所述预测模型不能够准确预测用户的患病风险，则该预测模型还需要进一步的训练。
步骤S305：若评分预测的结果在预设误差范围内，则确定预测模型为目标评分模型。
具体地，患病风险的预测装置101将剩余的1个测试集数据代入训练好的预测模型中进行患病的风险大小进行评分预测时，若评分预测的结果在预设误差范围内，则确定所述预测模型为所述目标评分模型，证明所述预测模型可以准确预测用户的患病风险。
可选的，患病风险的预测装置101还可以根据所述N个样本体检数据和所述目标评分模型，确定所述N个样本体检数据对应的风险分数，根据确定所述N个样本体检数据对应的风险分数，划分风险等级并确定所述风险分数与所述风险等级的映射关系。可以理解的是，风险分数越高，器对应的风险等级越高，用户可能会患得疾病的概率也越大。因此，根据该映射关系，可以确定目标用户的风险分数对应的风险等级，进而可以判断出目标用户患病的概率。
步骤S306：根据关键因素对目标用户的体检数据进行数据筛选并处理，获得风险数据。
具体地，患病风险的预测装置101基于Xgboost算法，在所述K个训练集上建立i个回归树，确定所述体检数据中影响所述风险数据的s个不同的关键因子，其中，s为大于1的正整数。所述关键因素包括所述s个不同的关键因子；所述根据所述关键因素对目标用户的体检数据进行数据筛选并处理，获得风险数据，包括：确定在所述s个关键因子的影响力值中所述影响力值最大的前j个关键因子对应的所述目标用户的体检数据，其中，j为大于1且小于s的正整数；对所述前j个关键因子对应的目标用户的体检数据进行数据筛选，获得所述风险数据。可以理解的是，可以在目标用户的体检数据中筛选出前j个关键因子对应的所述目标用户的体检数据，然后在进行数据处理后获得最终的风险数据，该风险数据可以是按照正则表达方式处理为0-1特征数据或者将文字数据转化为的数值数据。
步骤S307：根据风险数据通过目标评分模型，确定目标用户的风险分数。
步骤S308：根据风险分数确定目标用户患病风险的风险等级。
具体地，步骤S307-步骤S308的实现还可以对应参照上述图2A中步骤S202-步骤S203的相关描述，此处不再赘述。
实施本申请实施例，可以首先对用户的体检数据进行数据处理，使得体检数据为数值形式，以便计算，然后将处理好的用户数据输入到风险预测模型中，风险预测模型预测该用户患病的风险大小。其中，该风险预测模型是基于海量的样本数据利用单变量分析相关性或决策树确定样本数据中影响Y特征数据的关键因子，再将其目标关键因子以及根据K-Folds交叉验证算法以及N个样本体检数据，得到的K个训练集及测试集输入xgboost算法建立的“三高”疾病风险预测模型中训练得到的预测模型，其次可以训练好的风险评分模型确定目标用户患病的风险大小，并且可以发现一些“三高”疾病的内在规律，控制疾病影响因素，预防疾病形成，或者抑制病情发展，为解决国家医保控费难题，提升医疗卫生服务水平提供AI支撑。
上述详细阐述了本申请实施例的方法，下面提供了与本申请实施例的相关患病风险的预测装置，患病风险的预测装置10可以是一种通过获取、处理、分析和提取有价值的数据，为第三方使用带来各种便利的服务设备。请参考附图4，图4是本申请实施例提供的一种患病风险的预测装置的结构示意图。可以包括接收单元401，处理单元402，第一确定单元403，第二确定单元404，还可以包括：模型训练单元405，第三确定单元406，第四确定单元407。
接收单元401，用于接收关键因素，所述关键因素用于指示目标疾病的疾病类型或指示影响所述目标疾病的影响因子。
处理单元402，用于根据所述关键因素对目标用户的体检数据进行数据筛选并处理，获得风险数据，所述风险数据包括血压数据、血脂数据以及血糖数据中的一种或多种。
第一确定单元403，用于根据所述风险数据通过目标评分模型，确定所述目标用户的风险分数，所述目标评分模型为基于Xgboost算法和K-Folds交叉验证算法装置训练出的预测模型。
第二确定单元404，用于根据所述风险分数确定所述目标用户患病风险的风险等级。
在一种可能实现的方式中，所述体检数据包括外科基本检查数据、内科基本检查数据、口腔检查数据、超声数据、心电图数据、眼科检查数据、妇科检查数据、生化检查数据、血压测量结果、血脂测量结果以及血糖测量结果中的一种或多种。
在一种可能实现的方式中，所述关键因素包括M个关键字，所述M个关键字用于指示所述目标疾病的疾病类型，其中，M为大于或等于1的正整数；所述处理单元402具体用于：根据所述M个关键字对所述体检数据进行数据筛选，确定筛选出的与所述M个关键字对应的体检数据为所述风险数据。
在一种可能实现的方式中，所述关键因素包括M个关键字，所述M个关键字用于指示所述目标疾病的疾病类型，其中，M为大于或等于1的正整数；所述处理单元402具体用于：将所述M个关键字对应的所述体检数据按照正则表达方式处理为0-1特征数据，获得所述风险数据。
在一种可能实现的方式中，所述装置还包括：模型训练单元405，用于根据所述风险数据通过目标评分模型，确定所述目标用户的风险分数之前，获取N个用户的体检数据为样本体检数据，其中，N个用户不包括所述目标用户，N为大于1的正整数；根据所述K-Folds交叉验证算法将获取的N个样本体检数据分成K-1个训练集及1个测试集，其中，K为大于1的正整数；基于Xgboost算法，在所述K-1个训练集上建立i个回归树训练预测模型，所述预测模型用于对患病的风险大小进行评分预测，其中，i为大于1的正整数；根据所述1个测试集测试，通过训练后的所述预测模型对患病的风险大小进行评分预测；若评分预测的结果在预设误差范围内，则确定所述预测模型为所述目标评分模型。
在一种可能实现的方式中，所述装置还包括：第三确定单元406，用于基于Xgboost算法，在所述K个训练集上建立i个回归树，确定所述体检数据中影响所述风险数据的s个不同的关键因子，其中，s为大于1的正整数；所述关键因素包括所述s个不同的关键因子；所述处理单元402具体用于：确定在所述s个关键因子的影响力值中所述影响力值最大的前j个关键因子对应的所述目标用户的体检数据，其中，j为大于1且小于s的正整数；对所述前j个关键因子对应的目标用户的体检数据进行数据筛选，获得所述风险数据。
在一种可能实现的方式中，所述装置还包括：第四确定单元407，用于根据所述N个样本体检数据和所述目标评分模型，确定所述N个样本体检数据对应的风险分数；根据确定所述N个样本体检数据对应的风险分数，划分风险等级并确定所述风险分数与所述风险等级的映射关系。
需要说明的是，各个操作的实现还可以对应参照图2A-图3所示的方法实施例的相应描述，此处不再赘述。
如图5所示，图5是本申请实施例提供的另一种患病风险的预测装置的结构示意图，该装置20包括至少一个处理器501，至少一个存储器502、至少一个通信接口503。此外，该设备还可以包括天线等通用部件，在此不再详述。
处理器501可以是通用中央处理器（CPU），微处理器，特定应用集成电路（application-specific integrated circuit，ASIC)，或一个或多个用于控制以上方案程序执行的集成电路。
通信接口503，用于与其他设备或通信网络通信，如以太网，无线接入网（RAN），核心网，无线局域网(Wireless Local Area Networks，WLAN)等。
存储器502可以是只读存储器（read-only memory，ROM) 或可存储静态信息和指令的其他类型的静态存储设备，随机存取存储器（random access memory，RAM) 或者可存储信息和指令的其他类型的动态存储设备，也可以是电可擦可编程只读存储器（Electrically Erasable Programmable Read-Only Memory，EEPROM）、只读光盘（Compact Disc Read-Only Memory，CD-ROM）或其他光盘存储、光碟存储（包括压缩光碟、激光碟、光碟、数字通用光碟、蓝光光碟等）、磁盘存储介质或者其他磁存储设备、或者能够用于携带或存储具有指令或数据结构形式的期望的程序代码并能够由计算机存取的任何其他介质，但不限于此。存储器可以是独立存在，通过总线与处理器相连接。存储器也可以和处理器集成在一起。
其中，所述存储器502用于存储执行以上方案的应用程序代码，并由处理器501来控制执行。所述处理器501用于执行所述存储器502中存储的应用程序代码。
存储器502存储的代码可执行以上图2A-图3提供的患病风险的预测方法，比如，当装置20为患病风险的预测101时，可以接收关键因素，根据所述关键因素对目标用户的体检数据进行数据筛选并处理，获得风险数据，所述风险数据包括血压数据、血脂数据以及血糖数据中的一种或多种；根据所述风险数据通过目标评分模型，确定所述目标用户的风险分数，所述目标评分模型为基于Xgboost算法和K-Folds交叉验证算法方法训练出的预测模型；根据所述风险分数确定所述目标用户患病风险的风险等级。
需要说明的是，本申请实施例中所描述的患病风险的预测装置20中各功能单元的功能可参照图2A-图3所示的方法实施例的相应描述，此处不再赘述。
本申请实施例还提供了一种计算机可读存储介质，其上存储有计算机程序，计算机程序被处理器执行时实现上述实施例中方法的步骤，或者，计算机程序被处理器执行时实现上述实施例中装置的各模块/单元的功能，这里不再赘述。可选的，本申请涉及的存储介质如计算机可读存储介质可以是非易失性的，也可以是易失性的。
在本申请中，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组件可以集成在一个组件也可以是各个组件单独物理存在，也可以是两个或两个以上组件集成在一个组件中。上述集成的组件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。
所述集成的组件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取存储介质中。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) 执行各个本申请实施例所述方法的全部或部分步骤。而前述的存储介质包括：U 盘、移动硬盘、只读存储器 (Read-Only Memory，ROM)、随机存取存储器 (Random Access Memory，RAM)、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。
应理解，在本申请的各种实施例中，上述各过程的序号的大小并不意味着执行顺序的先后，各过程的执行顺序应以其功能和内在逻辑确定，而不应对本申请实施例的实施过程构成任何限定。尽管在此结合各实施例对本申请进行了描述，然而，在实施例所要求保护的本申请过程中，本领域技术人员可理解并实现公开实施例的其他变化。

Claims (20)

  1. 一种患病风险的预测方法，其中，包括：
    接收关键因素，所述关键因素用于指示目标疾病的疾病类型或指示影响所述目标疾病的影响因子；
    根据所述关键因素对目标用户的体检数据进行数据筛选并处理，获得风险数据，所述风险数据包括血压数据、血脂数据以及血糖数据中的一种或多种；
    根据所述风险数据通过目标评分模型，确定所述目标用户的风险分数，所述目标评分模型为基于Xgboost算法和K-Folds交叉验证算法方法训练出的预测模型；
    根据所述风险分数确定所述目标用户患病风险的风险等级。
  2. 根据权利要求1所述方法，其中，所述体检数据包括外科基本检查数据、内科基本检查数据、口腔检查数据、超声数据、心电图数据、眼科检查数据、妇科检查数据、生化检查数据、血压测量结果、血脂测量结果以及血糖测量结果中的一种或多种。
  3. 根据权利要求1所述方法，其中，所述关键因素包括M个关键字，所述M个关键字用于指示所述目标疾病的疾病类型，其中，M为大于或等于1的正整数；
    所述根据所述关键因素对目标用户的体检数据进行数据筛选并处理，获得风险数据，包括：
    根据所述M个关键字对所述体检数据进行数据筛选，确定筛选出的与所述M个关键字对应的体检数据为所述风险数据。
  4. 根据权利要求1所述方法，其中，所述关键因素包括M个关键字，所述M个关键字用于指示所述目标疾病的疾病类型，其中，M为大于或等于1的正整数；
    所述根据所述关键因素对目标用户的体检数据进行数据筛选并处理，获得风险数据，包括：
    将所述M个关键字对应的所述体检数据按照正则表达方式处理为0-1特征数据，获得所述风险数据。
  5. 根据权利要求1所述方法，其中，所述根据所述风险数据通过目标评分模型，确定所述目标用户的风险分数之前，还包括：
    获取N个用户的体检数据为样本体检数据，其中，N个用户不包括所述目标用户，N为大于1的正整数；
    根据所述K-Folds交叉验证算法将获取的N个样本体检数据分成K-1个训练集及1个测试集，其中，K为大于1的正整数；
    基于Xgboost算法，在所述K-1个训练集上建立i个回归树训练预测模型，所述预测模型用于对患病的风险大小进行评分预测，其中，i为大于1的正整数；
    根据所述1个测试集测试，通过训练后的所述预测模型对患病的风险大小进行评分预测；
    若评分预测的结果在预设误差范围内，则确定所述预测模型为所述目标评分模型。
  6. 根据权利要求5所述方法，其中，所述方法还包括：
    基于Xgboost算法，在所述K个训练集上建立i个回归树，确定所述体检数据中影响所述风险数据的s个不同的关键因子，其中，s为大于1的正整数；
    所述关键因素包括所述s个不同的关键因子；所述根据所述关键因素对目标用户的体检数据进行数据筛选并处理，获得风险数据，包括：
    确定在所述s个关键因子的影响力值中所述影响力值最大的前j个关键因子对应的所述目标用户的体检数据，其中，j为大于1且小于s的正整数；
    对所述前j个关键因子对应的目标用户的体检数据进行数据筛选，获得所述风险数据。
  7. 根据权利要求5所述方法，其中，所述方法还包括：
    根据所述N个样本体检数据和所述目标评分模型，确定所述N个样本体检数据对应的风险分数；
    根据确定所述N个样本体检数据对应的风险分数，划分风险等级并确定所述风险分数与所述风险等级的映射关系。
  8. 一种患病风险的预测装置，其中，包括：
    接收单元，用于接收关键因素，所述关键因素用于指示目标疾病的疾病类型或指示影响所述目标疾病的影响因子；
    处理单元，用于根据所述关键因素对目标用户的体检数据进行数据筛选并处理，获得风险数据，所述风险数据包括血压数据、血脂数据以及血糖数据中的一种或多种；
    第一确定单元，用于根据所述风险数据通过目标评分模型，确定所述目标用户的风险分数，所述目标评分模型为基于Xgboost算法和K-Folds交叉验证算法方法训练出的预测模型；
    第二确定单元，用于根据所述风险分数确定所述目标用户患病风险的风险等级。
  9. 一种患病风险的预测装置，其中，包括处理组件、存储组件和通信模组件，处理组件、存储组件和通信组件相互连接，其中，存储组件用于存储计算机程序，通信组件用于与外部设备进行信息交互；处理组件被配置用于调用计算机程序，执行以下步骤：
    接收关键因素，所述关键因素用于指示目标疾病的疾病类型或指示影响所述目标疾病的影响因子；
    根据所述关键因素对目标用户的体检数据进行数据筛选并处理，获得风险数据，所述风险数据包括血压数据、血脂数据以及血糖数据中的一种或多种；
    根据所述风险数据通过目标评分模型，确定所述目标用户的风险分数，所述目标评分模型为基于Xgboost算法和K-Folds交叉验证算法方法训练出的预测模型；
    根据所述风险分数确定所述目标用户患病风险的风险等级。
  10. 根据权利要求9所述装置，其中，所述关键因素包括M个关键字，所述M个关键字用于指示所述目标疾病的疾病类型，其中，M为大于或等于1的正整数；
    所述处理组件在根据所述关键因素对目标用户的体检数据进行数据筛选并处理，获得风险数据时，具体用于执行以下步骤：
    根据所述M个关键字对所述体检数据进行数据筛选，确定筛选出的与所述M个关键字对应的体检数据为所述风险数据。
  11. 根据权利要求9所述装置，其中，所述关键因素包括M个关键字，所述M个关键字用于指示所述目标疾病的疾病类型，其中，M为大于或等于1的正整数；
    所述处理组件在根据所述关键因素对目标用户的体检数据进行数据筛选并处理，获得风险数据时，具体用于执行以下步骤：
    将所述M个关键字对应的所述体检数据按照正则表达方式处理为0-1特征数据，获得所述风险数据。
  12. 根据权利要求9所述装置，其中，所述根据所述风险数据通过目标评分模型，确定所述目标用户的风险分数之前，所述处理组件还用于执行以下步骤：
    获取N个用户的体检数据为样本体检数据，其中，N个用户不包括所述目标用户，N为大于1的正整数；
    根据所述K-Folds交叉验证算法将获取的N个样本体检数据分成K-1个训练集及1个测试集，其中，K为大于1的正整数；
    基于Xgboost算法，在所述K-1个训练集上建立i个回归树训练预测模型，所述预测模型用于对患病的风险大小进行评分预测，其中，i为大于1的正整数；
    根据所述1个测试集测试，通过训练后的所述预测模型对患病的风险大小进行评分预测；
    若评分预测的结果在预设误差范围内，则确定所述预测模型为所述目标评分模型。
  13. 根据权利要求12所述装置，其中，所述处理组件还用于执行以下步骤：
    基于Xgboost算法，在所述K个训练集上建立i个回归树，确定所述体检数据中影响所述风险数据的s个不同的关键因子，其中，s为大于1的正整数；
    所述关键因素包括所述s个不同的关键因子；所述处理组件在根据所述关键因素对目标用户的体检数据进行数据筛选并处理，获得风险数据时，具体用于执行以下步骤：
    确定在所述s个关键因子的影响力值中所述影响力值最大的前j个关键因子对应的所述目标用户的体检数据，其中，j为大于1且小于s的正整数；
    对所述前j个关键因子对应的目标用户的体检数据进行数据筛选，获得所述风险数据。
  14. 根据权利要求12所述装置，其中，所述处理组件还用于执行以下步骤：
    根据所述N个样本体检数据和所述目标评分模型，确定所述N个样本体检数据对应的风险分数；
    根据确定所述N个样本体检数据对应的风险分数，划分风险等级并确定所述风险分数与所述风险等级的映射关系。
  15. 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，所述计算机程序被处理器执行以实现以下步骤：
    接收关键因素，所述关键因素用于指示目标疾病的疾病类型或指示影响所述目标疾病的影响因子；
    根据所述关键因素对目标用户的体检数据进行数据筛选并处理，获得风险数据，所述风险数据包括血压数据、血脂数据以及血糖数据中的一种或多种；
    根据所述风险数据通过目标评分模型，确定所述目标用户的风险分数，所述目标评分模型为基于Xgboost算法和K-Folds交叉验证算法计算机可读存储介质训练出的预测模型；
    根据所述风险分数确定所述目标用户患病风险的风险等级。
  16. 根据权利要求15所述计算机可读存储介质，其中，所述关键因素包括M个关键字，所述M个关键字用于指示所述目标疾病的疾病类型，其中，M为大于或等于1的正整数；
    所述根据所述关键因素对目标用户的体检数据进行数据筛选并处理，获得风险数据时，具体实现以下步骤：
    根据所述M个关键字对所述体检数据进行数据筛选，确定筛选出的与所述M个关键字对应的体检数据为所述风险数据。
  17. 根据权利要求15所述计算机可读存储介质，其中，所述关键因素包括M个关键字，所述M个关键字用于指示所述目标疾病的疾病类型，其中，M为大于或等于1的正整数；
    所述根据所述关键因素对目标用户的体检数据进行数据筛选并处理，获得风险数据时，具体实现以下步骤：
    将所述M个关键字对应的所述体检数据按照正则表达方式处理为0-1特征数据，获得所述风险数据。
  18. 根据权利要求15所述计算机可读存储介质，其中，所述根据所述风险数据通过目标评分模型，确定所述目标用户的风险分数之前，所述计算机程序被处理器执行还用于实现以下步骤：
    获取N个用户的体检数据为样本体检数据，其中，N个用户不包括所述目标用户，N为大于1的正整数；
    根据所述K-Folds交叉验证算法将获取的N个样本体检数据分成K-1个训练集及1个测试集，其中，K为大于1的正整数；
    基于Xgboost算法，在所述K-1个训练集上建立i个回归树训练预测模型，所述预测模型用于对患病的风险大小进行评分预测，其中，i为大于1的正整数；
    根据所述1个测试集测试，通过训练后的所述预测模型对患病的风险大小进行评分预测；
    若评分预测的结果在预设误差范围内，则确定所述预测模型为所述目标评分模型。
  19. 根据权利要求18所述计算机可读存储介质，其中，所述计算机程序被处理器执行还用于实现以下步骤：
    基于Xgboost算法，在所述K个训练集上建立i个回归树，确定所述体检数据中影响所述风险数据的s个不同的关键因子，其中，s为大于1的正整数；
    所述关键因素包括所述s个不同的关键因子；所述根据所述关键因素对目标用户的体检数据进行数据筛选并处理，获得风险数据时，具体实现以下步骤：
    确定在所述s个关键因子的影响力值中所述影响力值最大的前j个关键因子对应的所述目标用户的体检数据，其中，j为大于1且小于s的正整数；
    对所述前j个关键因子对应的目标用户的体检数据进行数据筛选，获得所述风险数据。
  20. 根据权利要求18所述计算机可读存储介质，其中，所述计算机程序被处理器执行还用于实现以下步骤：
    根据所述N个样本体检数据和所述目标评分模型，确定所述N个样本体检数据对应的风险分数；
    根据确定所述N个样本体检数据对应的风险分数，划分风险等级并确定所述风险分数与所述风险等级的映射关系。
PCT/CN2020/111662 2019-10-15 2020-08-27 一种患病风险的预测方法及装置 Ceased WO2021073263A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910980973.9A CN110838366B (zh) 2019-10-15 2019-10-15 一种患病风险的预测方法及装置
CN201910980973.9 2019-10-15

Publications (1)

Publication Number Publication Date
WO2021073263A1 true WO2021073263A1 (zh) 2021-04-22

Family

ID=69575222

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/111662 Ceased WO2021073263A1 (zh) 2019-10-15 2020-08-27 一种患病风险的预测方法及装置

Country Status (2)

Country Link
CN (1) CN110838366B (zh)
WO (1) WO2021073263A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113362953A (zh) * 2021-06-23 2021-09-07 平安国际智慧城市科技股份有限公司 身体风险预测方法、装置、计算机设备及存储介质
CN114708111A (zh) * 2022-04-15 2022-07-05 阳光人寿保险股份有限公司 疾病预测模型的交叉验证方法、装置及电子设备
CN120561820A (zh) * 2025-07-29 2025-08-29 中卫信软件股份有限公司 一种ai体检智能主检诊断方法及系统

Families Citing this family (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110838366B (zh) * 2019-10-15 2024-12-24 平安科技（深圳）有限公司 一种患病风险的预测方法及装置
CN111312390A (zh) * 2020-02-27 2020-06-19 复旦大学附属中山医院 一种乳腺癌心脏毒性的评估方法，系统及存储装置
CN111696662A (zh) * 2020-05-26 2020-09-22 平安科技（深圳）有限公司 疾病预测方法、装置及存储介质
CN111785380B (zh) * 2020-07-01 2022-12-02 医渡云（北京）技术有限公司 传染性疾病患病风险等级的预测方法及装置、介质、设备
CN114550921B (zh) * 2020-11-26 2025-05-06 华瑞同康生物技术（深圳）有限公司 一种基于li-rads分级的患病风险预测建模方法
CN115769239A (zh) * 2021-04-29 2023-03-07 京东方科技集团股份有限公司 一种疾病预测方法、装置、电子设备和计算机可读存储介质
CN115705928B (zh) * 2021-08-05 2026-03-03 华瑞同康生物技术（深圳）有限公司 基于细胞增殖标志物的tc患病风险预测建模方法
CN113689928B (zh) * 2021-08-24 2023-06-20 深圳平安智慧医健科技有限公司 保养及预防患病方案的推荐方法、装置、设备及存储介质
CN113643813B (zh) * 2021-08-30 2024-07-09 平安医疗健康管理股份有限公司 基于人工智能的慢病随访监管方法、装置及计算机设备
CN113724878B (zh) * 2021-08-31 2024-05-10 平安科技（深圳）有限公司 基于机器学习的医疗风险信息推送方法及装置
CN114550941B (zh) * 2022-01-28 2024-09-17 南京信息工程大学 基于狄克斯特拉算法的低冗余度动脉粥样硬化风险预测方法
CN116682565B (zh) * 2023-07-28 2023-11-10 济南蓝博电子技术有限公司 一种数字化医疗信息在线监测方法、终端及介质
CN118737474B (zh) * 2024-09-03 2024-11-05 西南医科大学 一种血管外科临床护理安全隐患分析方法
CN120690445B (zh) * 2024-11-21 2025-12-09 北京勤境医疗科技有限公司 基于大数据及体征监测的热射病患病风险评估方法

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108257675A (zh) * 2018-02-07 2018-07-06 平安科技（深圳）有限公司 慢阻肺发病风险预测方法、服务器及计算机可读存储介质
CN109785976A (zh) * 2018-12-11 2019-05-21 青岛中科慧康科技有限公司 一种基于Soft-Voting的痛风病分期预测系统
US20190192010A1 (en) * 2017-12-22 2019-06-27 Viraj Mane Detection of flu using thermal imaging
CN110838366A (zh) * 2019-10-15 2020-02-25 平安科技（深圳）有限公司 一种患病风险的预测方法及装置

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110202486A1 (en) * 2009-07-21 2011-08-18 Glenn Fung Healthcare Information Technology System for Predicting Development of Cardiovascular Conditions
US10535424B2 (en) * 2016-02-19 2020-01-14 International Business Machines Corporation Method for proactive comprehensive geriatric risk screening
CN106909769A (zh) * 2016-09-30 2017-06-30 马立明 大数据心血管疾病风险监控系统
CN107423560A (zh) * 2017-06-27 2017-12-01 天方创新（北京）信息技术有限公司 基于评分模型对二型糖尿病进行风险评分的方法及装置
CN107403072A (zh) * 2017-08-07 2017-11-28 北京工业大学 一种基于机器学习的2型糖尿病预测预警方法
CN109523118A (zh) * 2018-10-11 2019-03-26 平安科技（深圳）有限公司 风险数据筛选方法、装置、计算机设备和存储介质
CN109948680B (zh) * 2019-03-11 2021-06-11 合肥工业大学 病历数据的分类方法及系统

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190192010A1 (en) * 2017-12-22 2019-06-27 Viraj Mane Detection of flu using thermal imaging
CN108257675A (zh) * 2018-02-07 2018-07-06 平安科技（深圳）有限公司 慢阻肺发病风险预测方法、服务器及计算机可读存储介质
CN109785976A (zh) * 2018-12-11 2019-05-21 青岛中科慧康科技有限公司 一种基于Soft-Voting的痛风病分期预测系统
CN110838366A (zh) * 2019-10-15 2020-02-25 平安科技（深圳）有限公司 一种患病风险的预测方法及装置

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113362953A (zh) * 2021-06-23 2021-09-07 平安国际智慧城市科技股份有限公司 身体风险预测方法、装置、计算机设备及存储介质
CN114708111A (zh) * 2022-04-15 2022-07-05 阳光人寿保险股份有限公司 疾病预测模型的交叉验证方法、装置及电子设备
CN114708111B (zh) * 2022-04-15 2025-03-14 阳光人寿保险股份有限公司 疾病预测模型的交叉验证方法、装置及电子设备
CN120561820A (zh) * 2025-07-29 2025-08-29 中卫信软件股份有限公司 一种ai体检智能主检诊断方法及系统

Also Published As

Publication number Publication date
CN110838366B (zh) 2024-12-24
CN110838366A (zh) 2020-02-25

Similar Documents

Publication Publication Date Title
WO2021073263A1 (zh) 一种患病风险的预测方法及装置
Nazer et al. Bias in artificial intelligence algorithms and recommendations for mitigation
Leisman et al. Development and reporting of prediction models: guidance for authors from editors of respiratory, sleep, and critical care journals
Guo et al. Socioeconomic disparities in eHealth literacy and preventive behaviors during the COVID-19 pandemic in Hong Kong: cross-sectional study
Ogundimu et al. Adequate sample size for developing prediction models is not simply related to events per variable
Bayer et al. An early warning scoring system to identify septic patients in the prehospital setting: the PRESEP score
Yun et al. Prediction of critical care outcome for adult patients presenting to emergency department using initial triage information: an XGBoost algorithm analysis
Harrison et al. A new risk prediction model for critical care: the Intensive Care National Audit & Research Centre (ICNARC) model
Steyerberg et al. Assessing the incremental value of diagnostic and prognostic markers: a review and illustration
US20250273333A1 (en) Methods and Apparatus for Diagnosis of Progressive Kidney Function Decline Using a Machine Learning Model
Uchmanowicz et al. Frailty syndrome and self-care ability in elderly patients with heart failure
CN111653359B (zh) 一种出血性疾病的智能预测模型构建方法和预测系统
CN111695593A (zh) 基于XGBoost的数据分类方法、装置、计算机设备及存储介质
CN107658023A (zh) 疾病预测方法、装置、介质和电子设备
Sadatsafavi et al. Model-based ROC curve: examining the effect of case mix and model calibration on the ROC plot
Caneda et al. The reliability of the brief visuospatial memory test-revised in Brazilian multiple sclerosis patients
van Es et al. Improving the diagnostic management of upper extremity deep vein thrombosis
Hong et al. Semi‐supervised validation of multiple surrogate outcomes with application to electronic medical records phenotyping
KR20240134901A (ko) 이미지 분석 및 통찰력 생성
Fidecicchi et al. Empowering patient risk management strategies: validation the AHRQ elixhauser mortality index on Italian hospital administrative dataset
Ben Yehuda et al. Early detection of pulmonary embolism in a general patient population immediately upon hospital admission using machine learning to identify new, unidentified risk factors: model development study
García et al. Allostatic Load as a Short‐Term Prognostic and Predictive Marker
CN117936092A (zh) 疾病风险评估系统、方法及相关设备
Tripepi et al. Reappraisal in two European cohorts of the prognostic power of left ventricular mass index in chronic kidney failure
Fehr et al. A causal framework for assessing the transportability of clinical prediction models

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: 20877486

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: 20877486

Country of ref document: EP

Kind code of ref document: A1