CN116092681A - Method, system, electronic device and storage medium for determining health index score - Google Patents
Method, system, electronic device and storage medium for determining health index score Download PDFInfo
- Publication number
- CN116092681A CN116092681A CN202310309643.3A CN202310309643A CN116092681A CN 116092681 A CN116092681 A CN 116092681A CN 202310309643 A CN202310309643 A CN 202310309643A CN 116092681 A CN116092681 A CN 116092681A
- Authority
- CN
- China
- Prior art keywords
- user
- data
- health
- sample
- predicted
- 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.)
- Granted
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
-
- 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
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention provides a method, a system, electronic equipment and a storage medium for determining health index scores, which are used for generating user portrait data of a user to be predicted based on health data of the user to be predicted; extracting feature data to be processed corresponding to the specified health index from user portrait data of the user to be predicted; inputting the feature data to be processed into a disease risk prediction model for disease risk prediction, and obtaining a first disease probability of a user to be predicted suffering from a specified disease after a preset time period; determining a target age interval corresponding to the age of the user to be predicted; and inputting the feature data to be processed and the first illness probability into a health scoring model corresponding to the target age interval for scoring, so as to obtain a first health index score corresponding to the user to be predicted. According to the method, for users in different age intervals, the multi-dimensional health index data are processed by adopting a disease risk prediction model and a health score model corresponding to the ages of the users, so that accurate disease risk and health index scores are obtained.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a method, a system, electronic equipment and a storage medium for determining health index scores.
Background
With the development of internet medical information technology, electronic health record data, physical examination data, data detected by wearable equipment and the like are generated in large quantity, and mining analysis is performed based on the data to form user portrait data. The user profile data is utilized to generate a health index score that can help the user understand the health of the individual.
The current way to generate a health index score is: and extracting data such as blood pressure, blood sugar, BMI and the like of the user from the user portrait data for rough evaluation, thereby obtaining a health index score. However, the data such as blood pressure, blood sugar and BMI of the user only reflect part of the factors of the health condition of the user, and different types of users adopt the same scoring rule, so that the health index score obtained by current evaluation cannot accurately reflect the health condition of the user, and the accuracy of the health index score is low.
Disclosure of Invention
In view of the above, embodiments of the present invention provide a method, a system, an electronic device, and a storage medium for determining a health index score, so as to solve the problem that the accuracy of the health index score is low in the current method of generating the health index score.
In order to achieve the above object, the embodiment of the present invention provides the following technical solutions:
an embodiment of the present invention in a first aspect discloses a method for determining a health index score, the method comprising:
generating user portrait data of a user to be predicted based on health data of the user to be predicted;
extracting feature data to be processed corresponding to the specified health index from the user portrait data of the user to be predicted;
inputting the feature data to be processed into a preset disease risk prediction model for disease risk prediction, and predicting to obtain a first disease probability of the user to be predicted suffering from the specified disease after a preset time period, wherein the disease risk prediction model is obtained by training a first preset model based on user portrait data of the user with the first sample;
determining a target age interval corresponding to the age of the user to be predicted;
and inputting the feature data to be processed and the first illness probability into a health scoring model corresponding to the target age interval for scoring to obtain a first health index score corresponding to the user to be predicted, wherein the health scoring model is obtained by training a second preset model based on user portrait data of a second sample user with age in the target age interval.
Preferably, generating user portrait data of the user to be predicted based on health data of the user to be predicted includes:
performing natural language processing on health data of a user to be predicted;
extracting data which accords with a preset user portrait model from the health data of the user to be predicted which is subjected to natural language processing to form user portrait data of the user to be predicted.
Preferably, the process of training the first preset model based on the user portrait data of the first sample user to obtain the disease risk prediction model includes:
determining the disease time of a first sample user when the first sample user suffers from a specified disease for the first time according to the health data of the first sample user;
extracting first sample feature data corresponding to the specified health index before the preset duration from the disease time from the user portrait data of the first sample user;
and training a first preset model by using the first sample characteristic data until the first preset model converges to obtain a disease risk prediction model.
Preferably, the process of training the second preset model to obtain the health scoring model based on the user portrait data of the second sample user whose age is in the target age range includes:
Extracting second sample feature data corresponding to the specified health index from user portrait data of a second sample user with age in the target age interval, wherein the second sample feature data carries a scoring mark;
inputting the second sample characteristic data into the disease risk prediction model to predict disease risk, and predicting to obtain a second disease probability of the second sample user suffering from a specified disease after the preset time period;
and training a second preset model based on the second sample characteristic data and the second illness probability until the second preset model converges to obtain the health scoring model.
Preferably, the method further comprises:
extracting third sample characteristic data corresponding to the specified health index from user portrait data of a third sample user with the age in the target age interval;
inputting the third sample characteristic data into the disease risk prediction model to predict disease risk, and predicting to obtain a third disease probability of the third sample user suffering from a specified disease after the preset time period;
inputting the third sample characteristic data and the third illness probability into the health scoring model for scoring to obtain a second health index score corresponding to the third sample user;
Determining a third health index score corresponding to the third sample user by using a scoring rule corresponding to the target age interval and the health data of the third sample user;
and when the second health index score and the third health index score are in the same scoring interval, optimizing the health score model by using the user portrait data of the third sample user.
In a second aspect, an embodiment of the present invention discloses a system for determining a health index score, the system comprising:
the generation unit is used for generating user portrait data of the user to be predicted based on the health data of the user to be predicted;
the extraction unit is used for extracting the feature data to be processed corresponding to the specified health index from the user portrait data of the user to be predicted;
the prediction unit is used for inputting the feature data to be processed into a preset disease risk prediction model to perform disease risk prediction, and predicting to obtain a first disease probability of the user to be predicted having the specified disease after a preset time period, wherein the disease risk prediction model is obtained by training a first preset model based on user portrait data of the user with the first sample;
the determining unit is used for determining a target age interval corresponding to the age of the user to be predicted;
And the scoring unit is used for inputting the feature data to be processed and the first illness probability into a health scoring model corresponding to the target age interval to score so as to obtain a first health index score corresponding to the user to be predicted, wherein the health scoring model is obtained by training a second preset model based on user portrait data of a second sample user with age in the target age interval.
Preferably, the generating unit is specifically configured to: performing natural language processing on health data of a user to be predicted; extracting data which accords with a preset user portrait model from the health data of the user to be predicted which is subjected to natural language processing to form user portrait data of the user to be predicted.
Preferably, the prediction unit includes:
the determining module is used for determining the disease time of the first sample user when the first sample user suffers from the specified disease for the first time according to the health data of the first sample user;
the extraction module is used for extracting first sample characteristic data corresponding to the specified health index before the preset duration from the disease time from the user portrait data of the first sample user;
and the training module is used for training a first preset model by using the first sample characteristic data until the first preset model converges to obtain a disease risk prediction model.
A third aspect of the embodiment of the present invention discloses an electronic device, where the electronic device includes a processor and a memory, where the memory is configured to store program code and data for a method for determining a health index score, and the processor is configured to invoke program instructions in the memory to execute the method for determining a health index score disclosed in the first aspect of the embodiment of the present invention.
A fourth aspect of the embodiment of the present invention discloses a storage medium, where the storage medium includes a storage program, where when the program runs, the device where the storage medium is controlled to execute a method for determining a health index score disclosed in the first aspect of the embodiment of the present invention.
Based on the method, the system, the electronic device and the storage medium for determining the health index score provided by the embodiment of the invention, the method comprises the following steps: generating user portrait data of the user to be predicted based on the health data of the user to be predicted; extracting feature data to be processed corresponding to the specified health index from user portrait data of the user to be predicted; inputting the feature data to be processed into a disease risk prediction model for disease risk prediction, and predicting to obtain a first disease probability of a user to be predicted suffering from a specified disease after a preset time period; determining a target age interval corresponding to the age of the user to be predicted; and inputting the feature data to be processed and the first illness probability into a health scoring model corresponding to the target age interval for scoring, so as to obtain a first health index score corresponding to the user to be predicted. In the scheme, a disease risk prediction model is built, and a health scoring model corresponding to an age interval is built. And extracting the feature data to be processed corresponding to the specified health index from the user portrait data. And processing the feature data to be processed by using the disease risk prediction model to obtain a first disease probability of the user to be predicted, and processing the first disease probability and the feature data to be processed by using the health scoring model to obtain a first health index score of the user to be predicted. Aiming at users in different age intervals, a disease risk prediction model and a health scoring model corresponding to the ages of the users are adopted to process multidimensional health index data, so that accurate disease risk and health index scores are obtained.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present invention, and that other drawings can be obtained according to the provided drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method for determining a health index score according to an embodiment of the present invention;
FIG. 2 is a flowchart of training to obtain a disease risk prediction model and a health scoring model according to an embodiment of the present invention;
FIG. 3 is a block diagram of a system for determining health index scores according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. Based on the embodiments in the present invention, all other embodiments obtained by those skilled in the art without undue burden, all falling within the scope of the present invention.
In this application, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
As known from the background art, at present, data such as blood pressure, blood sugar, BMI and the like of a user are generally extracted from user portrait data to perform rough evaluation, so as to obtain a health index score. However, the data such as blood pressure, blood sugar and BMI of the user only reflect part of the factors of the health condition of the user, and different types of users adopt the same scoring rule, so that the health index score obtained by current evaluation cannot accurately reflect the health condition of the user, and the accuracy of the health index score is low.
Therefore, the embodiment of the invention provides a method, a system, electronic equipment and a storage medium for determining health index scores, and aims to construct a disease risk prediction model and a health score model corresponding to an age interval. And extracting the feature data to be processed corresponding to the specified health index from the user portrait data. And processing the feature data to be processed by using the disease risk prediction model to obtain a first disease probability of the user to be predicted, and processing the first disease probability and the feature data to be processed by using the health scoring model to obtain a first health index score of the user to be predicted. Aiming at users in different age intervals, a disease risk prediction model and a health scoring model corresponding to the ages of the users are adopted to process multidimensional health index data, so that accurate disease risk and health index scores are obtained.
Referring to fig. 1, a flowchart of a method for determining a health index score according to an embodiment of the present invention is shown, where the method includes:
step S101: user portrait data of the user to be predicted is generated based on the health data of the user to be predicted.
It should be noted that the health data includes, but is not limited to, electronic health records, physical examination data, and the like.
In the specific implementation process of step S101, user portrait data of the user to be predicted is generated based on the health data of the user to be predicted. Specifically, natural Language Processing (NLP) is carried out on the health data of the user to be predicted; extracting data which accords with a preset user portrait model from the health data of the user to be predicted which is processed by natural language so as to form user portrait data of the user to be predicted.
It will be appreciated that the natural language processing of health data specifically refers to: and extracting information from the free text in the health data, so as to structure unstructured data.
In some embodiments, a user image model is predefined; performing natural language processing on the health data of the user to be predicted; and storing the health data processed by the natural language according to fields defined by the user portrait model, thereby forming user portrait data of the user to be predicted.
Step S102: extracting the feature data to be processed corresponding to the specified health index from the user portrait data of the user to be predicted.
It should be noted that the specified health indicators include, but are not limited to: age, sex, occupation, illness, combined operation, family history, smoking and drinking conditions, blood pressure, blood sugar, heart rate, BMI, exercise, life habit and other indexes affecting health.
In the specific implementation process of step S102, data corresponding to each specified health index is extracted from user portrait data of the user to be predicted as feature data to be processed.
Step S103: and inputting the feature data to be processed into a preset disease risk prediction model to predict the disease risk, and predicting to obtain a first disease probability of the user to be predicted suffering from the specified disease after the preset time.
It should be noted that, the disease risk prediction model is obtained by training a first preset model based on user portrait data of a first sample user, the first preset model may be a deep learning multitask model, a fitting mathematical model or an empirical model, and meanwhile, the first preset model may be another type of model, and the type of the first preset model is not limited in the scheme.
It is further noted that the first sample user may be a user suffering from a specified condition; the specified conditions include, but are not limited to: critical diseases such as apoplexy, cerebral hemorrhage, cerebral infarction, ischemic heart disease, high pressure heart disease, chronic obstructive pulmonary disease, lung cancer, hepatocarcinoma, gastric cancer, hypertension, diabetes, breast cancer, esophageal cancer, rectal cancer, atherosclerosis, and myocardial infarction.
The first preset model is trained through the user portrait data of each first sample user to obtain a disease risk prediction model, and then the risk of the user suffering from the specified disease can be predicted (represented by probability values) by utilizing the disease risk prediction model obtained through training, and the process of how to obtain the disease risk prediction model through training is explained first.
Description of training the first preset model to get a disease risk prediction model:
determining the disease time of the first sample user when the first sample user suffers from the specified disease for the first time according to the health data of the first sample user; specifically, based on the time of diagnosis of the disease and the diagnosis of the disease in the electronic health record of the first sample user, it is determined which specified disorder the first sample user suffers from and the time of onset of the specified disorder, which is used to characterize when the first sample user suffers from the specified disorder for the first time.
And extracting first sample characteristic data corresponding to the specified health index before the preset duration of the illness time from the user portrait data of the first sample user. "before a preset period of time from the time of illness" specifically means: the affected time is advanced for a preset period of time.
For example: setting preset time length to be 3 years and 5 years; and extracting first sample characteristic data corresponding to each specified health index 3 years away from the illness time from the user portrait data of the first sample user, and extracting first sample characteristic data corresponding to each specified health index 5 years away from the illness time.
The user portrait data of the first sample user is generated based on the health data of the first sample user; in detail, how to generate the user portrait data of the first sample user may refer to the related content of the user portrait data of the user to be predicted generated in the step S101, which is not described herein.
And training a first preset model by using the extracted first sample characteristic data of each first sample user until the first preset model converges to obtain a disease risk prediction model. Wherein the training goal is to predict the probability of suffering from a given condition after a predetermined length of time in the future.
The disease risk prediction model is used for predicting the risk of a certain user suffering from each specified disorder after a preset period of time.
The above is a relevant description of how to train to get a disease risk prediction model.
In the specific implementation process of step S103, the feature data to be processed is input into a disease risk prediction model to perform disease risk prediction, so as to obtain a first disease probability of the user to be predicted having the specified disease after the preset time.
For example: setting preset time length to be 3 years and 5 years; inputting the feature data to be processed of the user to be predicted into a disease risk prediction model for disease risk prediction, predicting to obtain a first disease probability that the user to be predicted has the specified disease after 3 years, and predicting to obtain a first disease probability that the user to be predicted has the specified disease after 5 years.
Step S104: and determining a target age interval corresponding to the age of the user to be predicted.
It is understood that different age intervals are divided for different people groups; for example: the method comprises the steps of dividing age intervals corresponding to various groups of people for the elderly (more than 70 years), neonatal (less than 28 days), pediatric and adult groups respectively; an age interval corresponds to a certain group of people.
Also for example: aiming at the aged people, infant people and adult people, the age intervals corresponding to the various people are respectively divided.
In the specific implementation step S104, determining a target age interval corresponding to the age of the user to be predicted, or determining an age interval covering the age of the user to be predicted, wherein the age interval covering the age of the user to be predicted is the target age interval; the target age interval may characterize which group of people the user to be predicted belongs to.
Step S105: and inputting the feature data to be processed and the first illness probability into a health scoring model corresponding to the target age interval for scoring, so as to obtain a first health index score corresponding to the user to be predicted.
It should be noted that, for different age intervals, the health score models corresponding to the age intervals are respectively obtained through training (that is, the health score models are obtained through training for each group of people, in other words, a plurality of health score models exist in the technical scheme described in the application at the same time). The health scoring model corresponding to the target age interval is obtained by training a second preset model based on user portrait data of a second sample user whose age is in the target age interval. The health scoring models corresponding to other age intervals are obtained by training a second preset model based on user portrait data of sample users with ages in other age intervals. Taking the process of training to obtain the health scoring model corresponding to the target age interval as an example, explanation is made on how to train to obtain the health scoring model.
The training is described in terms of a health scoring model corresponding to the target age interval:
and extracting second sample feature data corresponding to the specified health index from user portrait data of a second sample user with age in the target age interval, wherein the second sample feature data carries a scoring label which is set by a medical expert based on the user portrait data of the second sample user.
It should be noted that, the user portrait data of the second sample user is generated based on the health data of the second sample user; in detail, how to generate the user portrait data of the second sample user can refer to the related content of the user portrait data of the user to be predicted generated in the step S101, which is not described herein.
And inputting the second sample characteristic data into a trained disease risk prediction model to predict the disease risk, and predicting to obtain a second disease probability of the second sample user suffering from the specified disease after the preset time.
And training a second preset model based on the second sample characteristic data and the second illness probability until the second preset model converges to obtain a health scoring model corresponding to the target age interval. The second preset model may be a deep learning model, a fitting mathematical model, an empirical model, and other models, and the type of the second preset model is not specifically limited in this scheme.
It can be understood that, because the data size of the user portrait data of the second sample user is larger, and the medical expert has limited energy and time, the medical expert can only set scoring marks on part of the user portrait data of the second sample user, so that the data size of the extracted second sample feature data is limited, and the accuracy of the health scoring model obtained through training in the above manner can be further improved.
In some embodiments of the present invention, the way to improve the accuracy of the health scoring model is: from user profile data of a third sample user whose age is in the target age range, extracting third sample characteristic data corresponding to the specified health index; and inputting the third sample characteristic data into a disease risk prediction model to predict the disease risk, and predicting to obtain a third disease probability of the third sample user suffering from the specified disease after the preset time.
And inputting the third sample characteristic data and the third illness probability into a health scoring model corresponding to the target age interval for scoring, and obtaining a second health index score corresponding to the third sample user.
Corresponding scoring rules are set for different age intervals in advance, specifically, the designated health index of the user in each age interval and the disease probability of the user suffering from the designated disease after the preset time period are integrated, the scoring rules corresponding to each age interval are defined, and the scoring rules corresponding to each age interval can develop a corresponding rule engine.
Determining a third health index score corresponding to the third sample user by using a scoring rule corresponding to the target age interval and health data of the third sample user; specifically, on the health data of the third sample user, scoring is performed by using a rule engine corresponding to a scoring rule corresponding to the target age interval, so that a third health index score corresponding to the third sample user can be determined.
A plurality of scoring intervals are divided in advance, for example: dividing a plurality of scoring intervals of 0-20, 20-40, 40-60, 60-80, 80-100 and the like. When the second health index score and the third health index score are in the same scoring interval, the second health index score and the third health index score can be characterized to be consistent, and the health scoring model is optimized by using user portrait data of a third sample user, wherein the user portrait data of the third sample user is equivalent to incremental training data in the process of training the health scoring model.
When the second health index score and the third health index score are in different score intervals, the second health index score and the third health index score can be characterized to be inconsistent, and user portrait data of a third sample user can be sent to a medical expert for auditing; user profile data of a third sample user after the audit is passed can also be used to optimize the health score model.
After optimizing the health score model, the health index score of the user is determined by utilizing the optimized health score model, so that more accurate health index score is obtained.
The above is illustrative of training to obtain a health scoring model corresponding to a target age interval; through the training mode, medical professionals can train to obtain a health scoring model with higher accuracy only by marking part of sample characteristic data (setting scoring marking). The training process of the health score model corresponding to the other age interval can refer to the related content of the health score model corresponding to the training obtained target age interval, which is not described herein.
It should be noted that, to better understand how the scoring rule is used to determine the third health index score, how the scoring rule is used to determine the third health index score is explained by the following.
Setting the initial health index score of the third sample user to 100 points, and deducting the initial health index score of the third sample user according to the rule shown in the following A1 to A9, so as to obtain a third health index score corresponding to the third sample user.
A1, classifying each occupation according to the damage depth to the body health; and correspondingly deducting the initial health index score according to the occupation of the third sample user.
A2, according to the comprehensive consideration of sequelae of diseases, future survival rate of tumor diseases, influence on life quality of patients and the like, the diseases are classified into critical diseases (such as cardiovascular and cerebrovascular organic diseases, malignant tumors and the like), chronic diseases (hypertension, diabetes and the like), common diseases, light diseases (respiratory tract infection, myopia, ocular allergic diseases and the like) and emergency department common diseases and the like. According to the illness condition of the third sample user, combining the age interval of the age of the third sample user, giving different deduction rules to correspondingly deduct the initial health index score, and considering the worst condition if the third sample user has multiple concurrent diseases at the same time. For example: when the third sample user is in the elderly population (older than 70 years), the initial health index score is deducted by 80 points if the third sample user suffers from critical illness. If the third sample user is suffering from a critical illness when the third sample user is in the adult population, 60 points are deducted from the initial health index score, as this application is not limited.
A3, if critical diseases and chronic diseases exist in the past medical history of the third sample user, correspondingly deducting the initial health index score.
A4, if the third sample user performs the operation once, performing corresponding deduction on the initial health index score according to different conditions of important organ resection, heart chest operation, craniocerebral operation, tumor resection operation, breast cancer without metastasis during operation, appendectomy operation, liver partial resection operation, unilateral kidney and other non-important organ resection operation and the like.
A5, aiming at bad life habits such as smoking, drinking, staying up, eating greasy, sweet food and the like of the third sample user, the initial health index score is correspondingly deducted, and if no bad life habits exist, the score is not reduced.
A6, if the BMI, blood pressure, blood sugar and the like of the third sample user exceed the standard range in the target age interval, correspondingly deducting the initial health index score.
A7, deducting 5 points from the initial health index score if the third sample user has chemoradiotherapy conditions.
A8, deducting 5 points from the initial health index score if family personnel of the third sample user have history of malignant tumors, hypertension, diabetes, coronary heart disease and other chronic diseases.
A9, deducting 5 points from the initial health index score if the third sample user has old myocardial infarction and cerebral infarction (cerebral hemorrhage, cerebral trauma and cerebral thrombosis) sequelae in the past.
Weighted accumulation of the points of A1 to A9 may be satisfied with the third health index score being in the interval 0-100.
The above is a relevant description of scoring rules.
In the specific implementation process of step S105, the feature data to be processed and the first illness probability are input into a health scoring model corresponding to the target age interval for scoring, so as to obtain a first health index score corresponding to the user to be predicted.
In some embodiments, a first probability of illness of the user to be predicted having the specified condition after the preset time period is output, and a first health index score corresponding to the user to be predicted is output; the user to be predicted can know the probability of suffering from each appointed disease after the preset time through the first illness probability, and know the health condition of the user through the first health index score so as to take corresponding medical treatment measures in time.
In the embodiment of the invention, aiming at users in different age intervals, a disease risk prediction model and a health score model corresponding to the age of the user are adopted to process multi-dimensional health index data, so that accurate disease risk and health index scores are obtained.
To better explain how the disease risk prediction model and the health score model are trained, the training illustrated by the flowchart of fig. 2 for obtaining the disease risk prediction model and the health score model is illustrated, and fig. 2 includes the following steps:
step S201: and extracting first sample characteristic data corresponding to the specified health index before the preset duration of the illness time from the user portrait data of the first sample user.
Step S202: and training a first preset model by using the extracted first sample characteristic data of each first sample user until the first preset model converges to obtain a disease risk prediction model.
Step S203: and extracting second sample characteristic data corresponding to the specified health index from user portrait data of the second sample user with the age in the target age range. And setting scoring rules corresponding to the age intervals.
Step S204 and step S205 are performed for the second sample feature data; step S206 and step S207 are performed for the set scoring rule.
Step S204: and constructing a health scoring model corresponding to the target age interval by using the second sample characteristic data.
Step S205: and determining a second health index score corresponding to the third sample user through a health score model corresponding to the target age interval.
Step S206: and constructing rule engines corresponding to the age intervals according to scoring rules corresponding to the age intervals.
Step S207: and determining a third health index score corresponding to the third sample user by using a rule engine corresponding to the target age interval.
Step S208: and judging whether the second health index score and the third health index score are in the same scoring interval. When the second health index score and the third health index score are in the same scoring interval, executing step S210; when the second health index score and the third health index score are in different score intervals, step S209 is performed.
It should be noted that, when the second health index score and the third health index score are in the same score interval, the third health index score corresponding to the third sample user is characterized to be more definite, and at this time, the user portrait data of the third sample user can be used as incremental training data to optimize the health score model, so that the accuracy of the health score model can be further improved.
Step S209: and transmitting the user portrait data of the third sample user to a medical expert for auditing, and executing step S210 after the auditing is passed.
Step S210: the health score model is optimized using user profile data of the third sample user.
It should be noted that, the execution principle of each step in fig. 2 can be referred to the content of each step in fig. 1 in the above embodiment of the present invention, and the description thereof is omitted herein.
Corresponding to the method for determining a health index score provided in the above embodiment of the present invention, referring to fig. 3, the embodiment of the present invention further provides a structural block diagram of a system for determining a health index score, where the system includes: a generating unit 301, an extracting unit 302, a predicting unit 303, a determining unit 304, and a scoring unit 305;
the generating unit 301 is configured to generate user portrait data of the user to be predicted based on the health data of the user to be predicted.
In a specific implementation, the generating unit 301 is specifically configured to: performing natural language processing on health data of a user to be predicted; and extracting data conforming to a preset user portrayal model from the health data of the user to be predicted which is subjected to natural language processing to form user portrayal data of the user to be predicted.
And the extracting unit 302 is configured to extract, from user portrait data of the user to be predicted, feature data to be processed corresponding to the specified health index.
The prediction unit 303 is configured to input feature data to be processed into a preset disease risk prediction model for performing disease risk prediction, and predict to obtain a first disease probability that a user to be predicted has a specified disease after a preset time period, where the disease risk prediction model is obtained by training a first preset model based on user portrait data of the user with the first sample.
A determining unit 304, configured to determine a target age interval corresponding to the age of the user to be predicted.
The scoring unit 305 is configured to input the feature data to be processed and the first probability of illness into a health scoring model corresponding to the target age interval to score, and obtain a first health index score corresponding to the user to be predicted, where the health scoring model is obtained by training a second preset model based on user portrait data of a second sample user whose age is in the target age interval.
In the embodiment of the invention, aiming at users in different age intervals, a disease risk prediction model and a health score model corresponding to the age of the user are adopted to process multi-dimensional health index data, so that accurate disease risk and health index scores are obtained.
Preferably, in combination with the content shown in fig. 3, the prediction unit 303 includes: the device comprises a determining module, an extracting module and a training module; the execution principle of each module is as follows:
and the determining module is used for determining the disease time of the first sample user when the first sample user suffers from the specified disease for the first time according to the health data of the first sample user.
The extraction module is used for extracting first sample characteristic data corresponding to the specified health index before the preset duration of the illness time from the user portrait data of the first sample user.
And the training module is used for training the first preset model by using the first sample characteristic data until the first preset model converges to obtain a disease risk prediction model.
Preferably, in combination with the content shown in fig. 3, the scoring unit 305 includes: the device comprises a first extraction module, a first processing module and a training module; the execution principle of each module is as follows:
the first extraction module is used for extracting second sample feature data corresponding to the specified health index from user portrait data of a second sample user with age in the target age range, wherein the second sample feature data carries scoring labels.
The first processing module is used for inputting the second sample characteristic data into the disease risk prediction model to predict the disease risk, and predicting to obtain a second disease probability of the second sample user suffering from the specified disease after the preset time.
And the training module is used for training the second preset model based on the second sample characteristic data and the second illness probability until the second preset model converges to obtain a health scoring model.
Preferably, the scoring unit 305 further includes: the device comprises a second extraction module, a second processing module, a third processing module, a determination module and an optimization module; the execution principle of each module is as follows:
And the second extraction module is used for extracting third sample characteristic data corresponding to the specified health index from the user portrait data of the third sample user with the age in the target age range.
And the second processing module is used for inputting the third sample characteristic data into the disease risk prediction model to perform disease risk prediction, and predicting to obtain a third disease probability of the third sample user suffering from the specified disease after the preset time.
And the third processing module is used for inputting the third sample characteristic data and the third illness probability into a health scoring model for scoring to obtain a second health index score corresponding to the third sample user.
And the determining module is used for determining a third health index score corresponding to the third sample user by utilizing the scoring rule corresponding to the target age interval and the health data of the third sample user.
And the optimizing module is used for optimizing the health score model by using the user portrait data of the third sample user when the second health index score and the third health index score are in the same scoring interval.
Preferably, an embodiment of the present invention provides an electronic device, as shown in fig. 4, where the electronic device includes a processor 401 and a memory 402, where the memory 402 is configured to store program code and data for a method for determining a health index score, and where the processor 401 is configured to invoke program instructions in the memory 402 to perform steps shown in implementing the method for determining a health index score in the above embodiment.
The embodiment of the invention also provides a storage medium, which comprises a storage program, wherein the storage medium is controlled to be located in equipment to execute the method for determining the health index score shown in the embodiment when the program runs.
In summary, the embodiments of the present invention provide a method, a system, an electronic device, and a storage medium for determining a health index score, and construct a disease risk prediction model, and construct a health score model corresponding to an age interval. And extracting the feature data to be processed corresponding to the specified health index from the user portrait data. And processing the feature data to be processed by using the disease risk prediction model to obtain a first disease probability of the user to be predicted, and processing the first disease probability and the feature data to be processed by using the health scoring model to obtain a first health index score of the user to be predicted. Aiming at users in different age intervals, a disease risk prediction model and a health scoring model corresponding to the ages of the users are adopted to process multidimensional health index data, so that accurate disease risk and health index scores are obtained.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for a system or system embodiment, since it is substantially similar to a method embodiment, the description is relatively simple, with reference to the description of the method embodiment being made in part. The systems and system embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative elements and steps are described above generally in terms of functionality in order to clearly illustrate the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A method of determining a health index score, the method comprising:
generating user portrait data of a user to be predicted based on health data of the user to be predicted;
extracting feature data to be processed corresponding to the specified health index from the user portrait data of the user to be predicted;
inputting the feature data to be processed into a preset disease risk prediction model for disease risk prediction, and predicting to obtain a first disease probability of the user to be predicted suffering from the specified disease after a preset time period, wherein the disease risk prediction model is obtained by training a first preset model based on user portrait data of the user with the first sample;
determining a target age interval corresponding to the age of the user to be predicted;
and inputting the feature data to be processed and the first illness probability into a health scoring model corresponding to the target age interval for scoring to obtain a first health index score corresponding to the user to be predicted, wherein the health scoring model is obtained by training a second preset model based on user portrait data of a second sample user with age in the target age interval.
2. The method of claim 1, wherein generating user profile data for a user to be predicted based on health data for the user to be predicted comprises:
Performing natural language processing on health data of a user to be predicted;
extracting data which accords with a preset user portrait model from the health data of the user to be predicted which is subjected to natural language processing to form user portrait data of the user to be predicted.
3. The method of claim 1, wherein training the first predetermined model based on the user profile data of the first sample user to obtain the disease risk prediction model comprises:
determining the disease time of a first sample user when the first sample user suffers from a specified disease for the first time according to the health data of the first sample user;
extracting first sample feature data corresponding to the specified health index before the preset duration from the disease time from the user portrait data of the first sample user;
and training a first preset model by using the first sample characteristic data until the first preset model converges to obtain a disease risk prediction model.
4. The method of claim 1, wherein training a second predetermined model based on user profile data of a second sample user having an age within the target age interval to obtain a health score model comprises:
Extracting second sample feature data corresponding to the specified health index from user portrait data of a second sample user with age in the target age interval, wherein the second sample feature data carries a scoring mark;
inputting the second sample characteristic data into the disease risk prediction model to predict disease risk, and predicting to obtain a second disease probability of the second sample user suffering from a specified disease after the preset time period;
and training a second preset model based on the second sample characteristic data and the second illness probability until the second preset model converges to obtain the health scoring model.
5. The method according to claim 4, wherein the method further comprises:
extracting third sample characteristic data corresponding to the specified health index from user portrait data of a third sample user with the age in the target age interval;
inputting the third sample characteristic data into the disease risk prediction model to predict disease risk, and predicting to obtain a third disease probability of the third sample user suffering from a specified disease after the preset time period;
inputting the third sample characteristic data and the third illness probability into the health scoring model for scoring to obtain a second health index score corresponding to the third sample user;
Determining a third health index score corresponding to the third sample user by using a scoring rule corresponding to the target age interval and the health data of the third sample user;
and when the second health index score and the third health index score are in the same scoring interval, optimizing the health score model by using the user portrait data of the third sample user.
6. A system for determining a health index score, the system comprising:
the generation unit is used for generating user portrait data of the user to be predicted based on the health data of the user to be predicted;
the extraction unit is used for extracting the feature data to be processed corresponding to the specified health index from the user portrait data of the user to be predicted;
the prediction unit is used for inputting the feature data to be processed into a preset disease risk prediction model to perform disease risk prediction, and predicting to obtain a first disease probability of the user to be predicted having the specified disease after a preset time period, wherein the disease risk prediction model is obtained by training a first preset model based on user portrait data of the user with the first sample;
the determining unit is used for determining a target age interval corresponding to the age of the user to be predicted;
And the scoring unit is used for inputting the feature data to be processed and the first illness probability into a health scoring model corresponding to the target age interval to score so as to obtain a first health index score corresponding to the user to be predicted, wherein the health scoring model is obtained by training a second preset model based on user portrait data of a second sample user with age in the target age interval.
7. The system according to claim 6, wherein the generating unit is specifically configured to: performing natural language processing on health data of a user to be predicted; extracting data which accords with a preset user portrait model from the health data of the user to be predicted which is subjected to natural language processing to form user portrait data of the user to be predicted.
8. The system of claim 6, wherein the prediction unit comprises:
the determining module is used for determining the disease time of the first sample user when the first sample user suffers from the specified disease for the first time according to the health data of the first sample user;
the extraction module is used for extracting first sample characteristic data corresponding to the specified health index before the preset duration from the disease time from the user portrait data of the first sample user;
And the training module is used for training a first preset model by using the first sample characteristic data until the first preset model converges to obtain a disease risk prediction model.
9. An electronic device comprising a processor and a memory for storing program code and data for a method of determining a health index score, the processor for invoking program instructions in the memory to perform a method of determining a health index score as claimed in any of claims 1-5.
10. A storage medium comprising a stored program, wherein the program, when run, controls a device in which the storage medium is located to perform a method of determining a health index score according to any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310309643.3A CN116092681B (en) | 2023-03-28 | 2023-03-28 | Method, system, electronic device and storage medium for determining health index score |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310309643.3A CN116092681B (en) | 2023-03-28 | 2023-03-28 | Method, system, electronic device and storage medium for determining health index score |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116092681A true CN116092681A (en) | 2023-05-09 |
| CN116092681B CN116092681B (en) | 2023-08-08 |
Family
ID=86204798
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310309643.3A Active CN116092681B (en) | 2023-03-28 | 2023-03-28 | Method, system, electronic device and storage medium for determining health index score |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116092681B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117198513A (en) * | 2023-11-07 | 2023-12-08 | 北京烔凡科技有限公司 | Health guidance method and system for hypertension patient |
| CN117238510A (en) * | 2023-11-16 | 2023-12-15 | 天津中医药大学第二附属医院 | Sepsis prediction method and system based on deep learning |
| CN117351484A (en) * | 2023-10-12 | 2024-01-05 | 深圳市前海高新国际医疗管理有限公司 | Tumor stem cell characteristic extraction and classification system based on AI |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109712712A (en) * | 2018-12-13 | 2019-05-03 | 平安医疗健康管理股份有限公司 | A kind of health evaluating method, health evaluating device and computer readable storage medium |
| CN109785965A (en) * | 2018-12-13 | 2019-05-21 | 平安医疗健康管理股份有限公司 | A kind of health evaluating method, health evaluating device and computer readable storage medium |
| US20190355448A1 (en) * | 2016-06-28 | 2019-11-21 | Spot Check Medical Surgical Equipment & Instruments Trading | Automated health assessment system and method thereof |
| CN112185561A (en) * | 2020-09-28 | 2021-01-05 | 平安医疗健康管理股份有限公司 | User portrait generation method and device and computer equipment |
| US20220076841A1 (en) * | 2020-09-09 | 2022-03-10 | X-Act Science, Inc. | Predictive risk assessment in patient and health modeling |
-
2023
- 2023-03-28 CN CN202310309643.3A patent/CN116092681B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190355448A1 (en) * | 2016-06-28 | 2019-11-21 | Spot Check Medical Surgical Equipment & Instruments Trading | Automated health assessment system and method thereof |
| CN109712712A (en) * | 2018-12-13 | 2019-05-03 | 平安医疗健康管理股份有限公司 | A kind of health evaluating method, health evaluating device and computer readable storage medium |
| CN109785965A (en) * | 2018-12-13 | 2019-05-21 | 平安医疗健康管理股份有限公司 | A kind of health evaluating method, health evaluating device and computer readable storage medium |
| US20220076841A1 (en) * | 2020-09-09 | 2022-03-10 | X-Act Science, Inc. | Predictive risk assessment in patient and health modeling |
| CN112185561A (en) * | 2020-09-28 | 2021-01-05 | 平安医疗健康管理股份有限公司 | User portrait generation method and device and computer equipment |
Non-Patent Citations (2)
| Title |
|---|
| SUHYEON KIM: "Risk score-embedded deep learning for biological ageestimation: Development and validation", 《INFORMATION SCIENCES》, vol. 586, pages 628 - 643 * |
| 刘莉 等: "健康画像在慢阻肺个性化健康管理系统中的应用研究", 《中国医学物理学杂志》, vol. 37, no. 7, pages 918 - 926 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117351484A (en) * | 2023-10-12 | 2024-01-05 | 深圳市前海高新国际医疗管理有限公司 | Tumor stem cell characteristic extraction and classification system based on AI |
| CN117198513A (en) * | 2023-11-07 | 2023-12-08 | 北京烔凡科技有限公司 | Health guidance method and system for hypertension patient |
| CN117198513B (en) * | 2023-11-07 | 2024-03-29 | 北京烔凡科技有限公司 | Health guidance method and system for hypertension patient |
| CN117238510A (en) * | 2023-11-16 | 2023-12-15 | 天津中医药大学第二附属医院 | Sepsis prediction method and system based on deep learning |
| CN117238510B (en) * | 2023-11-16 | 2024-02-06 | 天津中医药大学第二附属医院 | Sepsis prediction method and system based on deep learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116092681B (en) | 2023-08-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116092681B (en) | Method, system, electronic device and storage medium for determining health index score | |
| Kwon et al. | Deep learning for predicting in‐hospital mortality among heart disease patients based on echocardiography | |
| Zhang et al. | Sleep stage classification based on multi-level feature learning and recurrent neural networks via wearable device | |
| CN108648827B (en) | Cardiovascular and cerebrovascular disease risk prediction method and device | |
| WO2019159007A1 (en) | A system and method for documenting a patient medical history | |
| Natarajan et al. | Measurement of respiratory rate using wearable devices and applications to COVID-19 detection | |
| Wen et al. | Accurate prognostic awareness and preference states influence the concordance between terminally ill cancer patients’ states of preferred and received life-sustaining treatments in the last 6 months of life | |
| CN115497616B (en) | Method, system, equipment and storage medium for auxiliary decision-making of infectious diseases | |
| Yun et al. | Prediction of critical care outcome for adult patients presenting to emergency department using initial triage information: an XGBoost algorithm analysis | |
| Dar et al. | Design and development of hybrid optimization enabled deep learning model for COVID-19 detection with comparative analysis with DCNN, BIAT-GRU, XGBoost | |
| Li et al. | Progress in biological age research | |
| Hadj-Amar et al. | Bayesian model search for nonstationary periodic time series | |
| CN116110582A (en) | Health risk assessment method based on pre-training and multitasking bidirectional regulation mechanism | |
| Han et al. | Systematic review of health state utility values used in European pharmacoeconomic evaluations for chronic hepatitis C: Impact on cost-effectiveness results | |
| CN112542242A (en) | Data transformation/symptom scoring | |
| Schafer et al. | Cohorts and emerging health disparities: Biomorphic trajectories in China, 1989 to 2006 | |
| CN111968740B (en) | Diagnostic label recommendation method and device, storage medium and electronic equipment | |
| Saleh et al. | Healthcare Embedded System for Predicting Parkinson's Disease Based on AI of Things | |
| CN111430037A (en) | Similar medical record searching method and system | |
| CN112259260A (en) | Intelligent medical question and answer method, system and device based on intelligent wearable equipment | |
| Fonseca et al. | A computationally efficient algorithm for wearable sleep staging in clinical populations | |
| Lemoine et al. | Detrended windowed (lag one) autocorrelation: A new method for distinguishing between event-based and emergent timing | |
| Korsakov et al. | Deep and machine learning models to improve risk prediction of cardiovascular disease using data extraction from electronic health records | |
| van Houwelingen et al. | Cox regression model | |
| CN113936767A (en) | Health management method for asthma patient |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |