CN113940640B - Cardiovascular disease risk control method, system and storage medium - Google Patents
Cardiovascular disease risk control method, system and storage medium Download PDFInfo
- Publication number
- CN113940640B CN113940640B CN202111339208.2A CN202111339208A CN113940640B CN 113940640 B CN113940640 B CN 113940640B CN 202111339208 A CN202111339208 A CN 202111339208A CN 113940640 B CN113940640 B CN 113940640B
- Authority
- CN
- China
- Prior art keywords
- cardiovascular disease
- disease risk
- risk control
- agent
- data
- 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.)
- Active
Links
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/02007—Evaluating blood vessel condition, e.g. elasticity, compliance
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
- A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7271—Specific aspects of physiological measurement analysis
- A61B5/7275—Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/74—Details of notification to user or communication with user or patient ; user input means
- A61B5/746—Alarms related to a physiological condition, e.g. details of setting alarm thresholds or avoiding false alarms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
Abstract
The invention relates to the field of intelligent medical treatment, and provides a cardiovascular disease risk control method, a cardiovascular disease risk control system and a storage medium, which can be used for early warning the cardiovascular disease risk of a patient to be detected and helping the patient with high cardiovascular disease risk to continuously adjust the living habits of the patient by providing a cardiovascular disease risk control scheme, so that the technical effect of reducing the future risk of suffering from cardiovascular diseases is achieved.
Description
Technical Field
The invention relates to the technical field of intelligent medical treatment, and relates to a cardiovascular disease risk control method, a cardiovascular disease risk control system, a cardiovascular disease risk control storage medium, a cardiovascular disease risk control device, an electronic device and a computer-readable storage medium.
Background
Cardiovascular diseases are the most common diseases in our country, and have been classified as the first killers endangering human health in the 21 st century by the world health organization. People with bad living habits such as staying up all night, smoking, sedentary and the like have significantly higher probability of suffering from cardiovascular diseases than people with healthy living habits. Thus, the risk of cardiovascular disease in an individual may be reduced by conducting a health management of cardiovascular disease.
The method for performing cardiovascular disease health management through a cardiovascular health management algorithm in the prior art has the following disadvantages:
1) The current indexes for cardiovascular disease risk assessment are only whether smoking is performed or not, cholesterol content, height, weight, age and the like; the evaluation indexes are limited, so that the evaluation result is poor in individuation and low in accuracy;
2) The given lifestyle adjustment scheme only has the target result of adjustment and does not contain reasonable steps of adjustment, so that the execution condition of the evaluated person is not ideal.
Therefore, there is a need for a health management method for reducing the risk of cardiovascular diseases that can embody individual differences.
Disclosure of Invention
The invention provides a cardiovascular disease risk control method, a cardiovascular disease risk control system, an electronic device and a computer-readable storage medium, which are used for overcoming the problems in the prior art and achieving the technical effect of reducing the risk of cardiovascular disease of a person to be detected by adjusting the living habits of the person to be detected by using a cardiovascular disease risk control agent.
In order to achieve the above object, the present invention provides a method for controlling cardiovascular disease risk, the method comprising: acquiring behavior data of a person to be detected, wherein the behavior data is individual state data at a certain time;
inputting the behavior data into a pre-trained cardiovascular disease risk prediction model to obtain a cardiovascular disease risk value of a person to be detected; screening the behavior data of the person to be detected with the cardiovascular disease risk value larger than a set risk threshold value as the state data of the person with high risk of cardiovascular disease;
inputting the state data into a pre-trained cardiovascular disease risk control intelligent agent to obtain a cardiovascular disease risk control scheme; wherein the cardiovascular disease risk control agent is obtained based on reinforcement learning training;
inputting the state data of the person to be detected after executing the cardiovascular disease risk control scheme into a cardiovascular disease risk control intelligent body, and obtaining a cardiovascular disease risk assessment value of the person to be detected in real time; the cardiovascular disease risk control agent determines a further cardiovascular disease risk control regimen based on the cardiovascular disease risk assessment value.
Further, preferably, the cardiovascular disease risk prediction model is:
f(x V )＝f(x DV ，x UV ，x IV )＝f(x DV ，x UV ，D(x DV ，x UV ))
wherein x is V To test the samples, f (x) V ) Is thatCardiovascular disease risk prediction model f vs x V A desired prediction output of; d is a regression model, the input of the regression model D is DV data and UV data, and the output is IV data;
DV is a directly variable feature, including daily smoking capacity, daily drinking capacity, and daily exercise capacity;
UV is an unchangeable characteristic including age, gender, and territory;
IV is an indirectly alterable feature including BMI index, blood glucose concentration, blood pressure, body fat rate and waist circumference.
Further, preferably, the training method of the cardiovascular disease risk control agent based on reinforcement learning comprises the following steps:
building an intelligent environment for cardiovascular disease risk control and a reinforcement learning model for cardiovascular disease risk control; wherein the agent environment is a group consisting of (S, A, R,
γ); wherein S is a state space comprising cardiovascular disease characteristics; cardiovascular disease characteristics are taken to include directly alterable, unalterable and indirectly alterable characteristics; a is an action space including directly alterable features, R is a reward function, and->
Is a state transition probability function, gamma ∈ (0,1)]Is a reward attenuation factor;
evaluating the sequence action output by the reinforcement learning model for cardiovascular disease risk control by using the intelligent agent environment, giving reward corresponding to the sequence action and the sequence action at the next moment, and performing centralized training by taking the maximization of accumulated reward as a target to obtain the pi of the intelligent agent * (as); wherein, the intelligent agent pi * The input of (a | s) is the state of the environment, and the output is the sequence action selectively executed in the state of the environment;
and iterating to the convergence of the intelligent agent model through the generalized dominance function and the strategy to obtain the trained cardiovascular disease risk control intelligent agent based on reinforcement learning.
Further, preferably, the cardiovascular disease risk prediction model is obtained by training an optimal characteristic variable set; the method for acquiring the optimal characteristic variable set comprises the following steps:
taking a data set containing cardiovascular disease behavior characteristics as a training set to input a behavior characteristic importance evaluation model; performing importance evaluation on the behavior characteristics of the cardiovascular diseases in the training set through a behavior characteristic importance evaluation model, and outputting the behavior characteristics of the cardiovascular diseases of which the importance evaluation accords with a set threshold value as an influence factor set;
and taking the image factor set as an optimal characteristic variable set.
Further, preferably, the jackpot is obtained by the following formula:
wherein, B(s) t ，a t ) For background score, B(s) t ，a t )＝α(c-f(s t +a t ) α is the reward scaling factor, c represents the margin factor for action, c ∈ (0,1)]，f(s t +a t ) Is in a state of s t The action is a t The output value of the cardiovascular disease risk prediction model; -p is a penalty score parameter,
divide the parameters for a terminal and when the agent wins >>
Intelligent case failing then->
When the intelligent agent is in a non-terminal state, then->
Further, preferably, the method for converging the intelligent agent model through the generalized dominance function and the strategy iteration includes:
estimation with merit function
And->
Wherein the content of the first and second substances,
is an estimate of the merit function at time step t, q π For the policy lifting function, v π Evaluating a function for the policy;
computing policy updates
After K time steps, strategy updating is executed by using small-batch SGD; wherein θ represents a parameter of the agent policy; />
Is a jackpot; pi (a | s) is a policy function; pi (a | s) is a policy function; />
Is the state transition probability; gamma is a reward attenuation factor; π θ (at/st) represents the expectation of a jackpot to select an action value at under a given policy π, state value st; />
Is a loss function;
and continuously iterating the steps until the intelligent agent model converges.
wherein E represents desired;
is the ratio of old and new strategies, where theta is the parameter of the neural network of the strategy, pi θ (a t |s t ) Represents the strategy at time t, < >>
The strategy of the previous iteration is adopted;
clip(r t (theta), 1-epsilon, 1+ epsilon) indicates that the proportion of the new strategy and the old strategy is controlled to be close to 1;
ε represents a hyperparameter.
In order to solve the above problems, the present invention also provides a cardiovascular disease risk control system,
the behavior data acquisition unit is used for acquiring behavior data of a person to be detected, wherein the behavior data is individual state data at a certain time;
the state data screening unit of the cardiovascular disease high-risk person is used for inputting the behavior data into a pre-trained cardiovascular disease risk prediction model to obtain a cardiovascular disease risk value of the person to be detected; screening the behavior data of the person to be detected with the cardiovascular disease risk value larger than a set risk threshold value as the state data of the person with high risk of cardiovascular disease;
the cardiovascular disease risk control scheme acquisition unit is used for inputting the state data into a pre-trained cardiovascular disease risk control intelligent agent to obtain a cardiovascular disease risk control scheme; wherein the cardiovascular disease risk control agent is obtained based on reinforcement learning training;
the cardiovascular disease risk control scheme adjusting unit is used for inputting the state data of the person to be detected after executing the cardiovascular disease risk control scheme into the cardiovascular disease risk control intelligent agent and obtaining the cardiovascular disease risk assessment value of the person to be detected in real time; the cardiovascular disease risk control agent determines a further cardiovascular disease risk control regimen based on the cardiovascular disease risk assessment value.
In order to solve the above problem, the present invention also provides an electronic device, including:
a memory storing at least one instruction; and
and the processor executes the instructions stored in the memory to realize the steps of the cardiovascular disease risk control method.
In order to solve the above problem, the present invention further provides a computer-readable storage medium, having at least one instruction stored therein, where the at least one instruction is executed by a processor in an electronic device to implement the cardiovascular disease risk control method described above.
The cardiovascular disease risk control method can warn the cardiovascular disease risk of a person to be detected, and can help the person with high cardiovascular disease risk to continuously adjust the living habits of the person with high cardiovascular disease risk by providing a cardiovascular disease risk control scheme, thereby achieving the technical effect of reducing the future cardiovascular disease risk.
Drawings
FIG. 1 is a schematic flow chart of a cardiovascular disease risk control method according to an embodiment of the present invention;
FIG. 2 is a graph illustrating the convergence of an agent for cardiovascular risk prediction provided by an embodiment of the present invention;
FIG. 3 is a comparative graph of cardiovascular disease risk reduction before and after lifestyle modification by a cardiovascular disease risk prediction method according to an embodiment of the present invention;
FIG. 4 is a graph showing the effect of reducing the risk of cardiovascular disease after 5-step lifestyle modification by the cardiovascular disease risk prediction method according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a system for risk control of cardiovascular diseases according to an embodiment of the present invention;
fig. 6 is a schematic diagram of an internal structure of an electronic device for implementing a cardiovascular disease risk control method according to an embodiment of the present invention;
the implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Referring to fig. 1, a schematic flow chart of a cardiovascular disease risk control method according to an embodiment of the present invention is shown. The method may be performed by an apparatus, which may be implemented by software and/or hardware.
In this embodiment, the cardiovascular disease risk control method includes S1 to S4:
s1, acquiring behavior data of a person to be detected, wherein the behavior data is individual state data at a certain time.
In particular implementations, the status data includes physiological data including, but not limited to, daily smoking volume, daily drinking volume, daily exercise amount, age, sex, region, family member status, BMI index, blood glucose concentration, blood pressure, systolic blood pressure, total Cholesterol (TC), high density lipoprotein cholesterol (HDL-C), whether to take hypotensive drugs, whether to have diabetes, body fat rate, sleep status, waist circumference, and prior related medical history. Of particular concern are high risk factors for cardiovascular disease, such as smoking, obesity, hypertension, diabetes, hyperlipidemia, family history of cardiovascular disease, and the like. Including but not limited to, whether to smoke, whether there is a family history of cardiovascular disease.
In a specific implementation process, the physiological data can be acquired through an intelligent body fat scale, an intelligent electrocardiograph, wearable equipment including a mobile phone, an intelligent bracelet and the like. For example, the person to be detected performs physiological status data acquisition through the smart band, and the physiological status data may include non-sleep status data and sleep status data, where the non-sleep status data includes an average heart rate, a maximum heart rate, a temperature value, a systolic pressure value, a diastolic pressure value, a blood pressure median, a blood oxygen value, and a blood lipid value. The sleep state data comprises resident sleep data including sleep time, sleep time data, deep sleep time and previous sleep time in the sleep process, real-time heart rate data, temperature values, systolic pressure values, diastolic pressure values, blood pressure median values, blood oxygen values and blood fat values. The intelligent bracelet utilizes WIFI or bluetooth to transmit physiological state data to APP in the mobile phone.
S2, inputting the behavior data into a pre-trained cardiovascular disease risk prediction model to obtain a cardiovascular disease risk value of a person to be detected; and screening the behavior data of the person to be detected with the cardiovascular disease risk value larger than a set risk threshold value as the state data of the person with high risk of cardiovascular disease.
For the cardiovascular disease risk prediction model, cardiovascular disease risk prediction labels are set according to historical data of cardiovascular diseases, and the cardiovascular disease risk prediction model based on machine learning or deep learning is trained.
The cardiovascular disease risk prediction model is as follows:
f(x V )＝f(x DV ，x UV ，x IV )＝f(x DV ，x UV ，D(x DV ，x UV ))
wherein x is V To test the samples, f (x) V ) Is cardiovascular disease risk prediction model f vs x V A desired prediction output of; d is a regression model, the input of the regression model D is DV data and UV data, and the output is IV data; DV is a directly variable feature including daily smoking volume, daily drinking volume, and daily exercise volume; UV is an unchangeable characteristic including age, sex and placeA domain; IV is an indirectly alterable feature including BMI index, blood glucose concentration, blood pressure, body fat rate and waist circumference.
It should be noted that the target feature to be managed is a directly changeable feature, and the indirectly changeable feature IV can be obtained by training the directly changeable feature and the indirectly unchangeable feature through a regression algorithm by using a deep learning technique, that is, y IV ＝D(x DV ，x UV ). The output of the cardiovascular disease risk prediction model is the risk (probability) of the individuals to be detected to suffer from cardiovascular diseases in the future. In particular implementations, the cardiovascular disease risk prediction model may employ, but is not limited to, realizable algorithms including SVM, random forest, rogers regression, deep neural networks, and the like.
The cardiovascular disease risk prediction model is obtained by training an optimal characteristic variable set; the method for obtaining the optimal characteristic variable set comprises the following steps: dividing a data set containing cardiovascular disease behavior characteristics into a training set, a verification set and a test set; inputting the training set into a behavior feature importance evaluation model; the behavior feature importance evaluation model carries out importance evaluation on the behavior features of the cardiovascular diseases in the training set, and the behavior features of the cardiovascular diseases with importance evaluation meeting a set threshold value are output as an influence factor set; and taking the image factor set as an optimal characteristic variable set.
Taking a random forest algorithm as an example, taking the training number set as the input of a random forest regression model to perform importance evaluation on behavior characteristics of cardiovascular disease risks, performing characteristic selection on influence factors according to the importance evaluation result, selecting an influence factor set with the minimum random forest regression model error, and taking the influence factor set as an optimal characteristic variable set. The method comprises the following specific steps: s11, constructing a random forest regression model according to the feature numbers contained in the binary tree nodes in the random forest regression model and the number of the decision trees; s12, taking the training number set as input of a random forest regression model, and calculating out-of-bag data errors of each decision tree in the random forest regression model by adopting out-of-bag data corresponding to the decision tree; s13, randomly taking a certain variable of all sample data in the out-of-bag data corresponding to the decision tree as a characteristic X, adding noise interference to the characteristic X, and then calculating the out-of-bag data error corresponding to the decision tree again; s14, constructing an importance calculation model, and performing variable importance evaluation on the feature X according to the importance calculation model; s15, repeating the steps S12 to S14 until the variable importance evaluations of all the variables in the training number set are output, then drawing a visual drawing of the variable importance evaluations, arranging the variable importance evaluations of all the variables in a descending order, and primarily screening the importance measurement of the variables according to the ordering result; and S16, successively removing the variables of the designated proportion from the variable set by using a recursive characteristic backward elimination method for the variable set obtained after the preliminary screening, removing one variable each time, comparing the error rates outside the bags corresponding to the remaining variables after removing the variable, taking the variable set with the minimum error rate as the optimal characteristic variable set, and determining the number of the optimal characteristics in the optimal characteristic variable set according to the error rates.
S3, inputting the state data into a pre-trained cardiovascular disease risk control intelligent agent to obtain a cardiovascular disease risk control scheme; wherein the cardiovascular disease risk control agent is obtained based on reinforcement learning training. Specifically, the cardiovascular disease risk control agent performs predictive risk assessment according to long-span historical data of cardiovascular detection, and further makes appropriate decisions according to prediction results and user characteristics. How the agent reacts and decides the action to be made according to the state of the environment; the strategy model provided by the reinforcement learning platform supports a deterministic model and a stochastic model, the deterministic model refers to a strategy which is fixed, and the purpose of reinforcement learning is to obtain an intelligent strategy which maximizes long-term rewards.
It should be noted that the state s of the environment when the input of the agent is a certain time t t The output of an agent is a sequential action, an action a in the state space t (ii) a While the agent receives the prize R at each step and aims to maximize the cumulative prize from start to end. The agent is trained through multiple iterations, which may include, but are not limited toTraining is carried out on the existing reinforcement learning algorithms such as PPO, A3C, doubleDQN and the like.
The training method of the cardiovascular disease risk control agent based on reinforcement learning comprises the following steps:
s31, building an intelligent environment for cardiovascular disease risk control and a reinforcement learning model for cardiovascular disease risk control; wherein the agent environment is a mixture of (S, A, R,
γ); s is a state space comprising cardiovascular disease features; wherein the cardiovascular disease characteristic is taken to include a directly alterable characteristic, an unalterable characteristic, and an indirectly alterable characteristic; a is an action space comprising directly alterable features, R is a reward function, and ` er `>
Is a state transition probability function, gamma ∈ (0,1)]Is the reward attenuation factor. In a specific implementation process, when an environment is built, different coefficients and expressions can be selected, and the example discloses only one method for building the environment.
That is, the behavior data of the person to be detected and the cardiovascular disease risk prediction model are used for building a training environment of the intelligent body. The main contents in the environment are: (the sum of S, A, R,
γ). Wherein S is a state set of the environment, i.e. a state space comprising cardiovascular disease features; a is the set of actions that an individual and agent can perform, i.e., the action space that includes directly changeable features; />
Is a reward function>
Is the state transition probability of the environment, i.e. the state transition probability function; gamma epsilon (0,1)]Is the reward attenuation factor.
In the environment, the environment state of the t-th step is s t E.s and contains all the individual states V (including alcohol consumption, age, BMI index, body fat rate, etc.). When the agent interacts with the environment, e.g. performs action a t Thereafter, the environmental state will transition probabilities according to the state
Is converted into s t+1 。
In a specific implementation, the motion space a is set to include only directly changeable features. At the t-th step, action a t For selecting a directly changeable feature DV i E.g., DV, by one unit or by one unit. The specific method comprises the following steps: initialization a t If the vector is a vector of all 0 of dimension | DV |, then let
Representing an increase of one unit over the ith feature,
representing a decrease of one unit over the ith feature. Constraint | | a t || 1 =1, i.e. only one feature can be changed per step. Examples are to reduce the smoking amount by 1, to reduce the drinking amount by 1, or to increase the exercise amount by 1.
Deterministic state transitions
Since the probability of taking action a at state s is only related to the current state s, and not to other elements. Set in state s t In time, the operation a is performed t State s at time t +1 t+1 Only by s t And a t And (6) determining. The transfer rule is: a is to t Zero-padding extension to vector a 'of dimension | DV + UV + IV' t So that s t+1 ＝s t +a′ t . At this time>
Taking PPO2 as an example, the process of training the reinforcement learning model for cardiovascular disease risk control by utilizing the established intelligent environment for cardiovascular disease risk control is as follows:
s32, evaluating the sequence action output by the reinforcement learning model for cardiovascular disease risk control by using the intelligent environment, giving reward corresponding to the sequence action of the reinforcement learning model for cardiovascular disease risk control and the sequence action at the next moment, and carrying out centralized training by taking the maximization of accumulated reward as a target to obtain the pi of the intelligent agent * (as); wherein, the intelligent agent pi * The input of (a | s) is the state of the environment, and the output is the sequence action selected to be executed in the state of the environment.
That is, the continuous non-sparse reward function Rt for each action at is modeled; and determining the executed action at when the current environment state is St, generating the next environment state St +1 after the executed action at acts on the human body of the person to be detected, obtaining the reward of a corresponding reward function Rt, and sequentially and circularly iterating to realize the maximum cumulative reward.
In a specific embodiment, the combination is as follows,
the jackpot is obtained by the following formula:
wherein, B(s) t ，a t ) For background score, B(s) t ，a t )＝α(c-f(s t +a t ) α is the reward scaling factor, c represents the coefficient of tolerance to action, c ∈ (0,1)]，f(s t +a t ) Is in a state of s t The action is a t The output value of the cardiovascular disease risk prediction model; -p is a penalty score parameter,
divides the parameter for the final round and->
Intelligent agent failing then>
When the intelligent agent is in a non-terminal state, then->
That is, it is determined whether to end the sequence of actions based on whether the agent is in a terminal state. The end state of the agent is represented by the end state flag Done. When Done =0, continue the loop; and if Done =1, entering the end state and exiting the loop. When the intelligent agent actively selects to terminate the decision or is terminated by the environment due to violation of the constraint condition to continue the operation, done =1; in the normal decision step, done is kept at 0.
It should be noted that, within the limited number of steps T, the current state s is enabled t And if the output of the cardiovascular disease risk prediction model is less than the threshold value C, the intelligent agent is considered to win. If the number of steps T is exceeded, the current state s is not enabled t If the output of the cardiovascular disease risk prediction model is less than the threshold value C, the agent fails. In a specific implementation, -p, T, α and
all parameters are set according to practical application scenarios, and examples are as follows, setting p =0.01, α =10, t =50,
and S33, iterating to the convergence of the intelligent agent model through the generalized dominant function and the strategy to obtain the trained cardiovascular disease risk control intelligent agent based on reinforcement learning.
The method for converging the intelligent agent model through the generalized dominance function and the strategy iteration comprises the following steps:
estimation with merit function
Is the estimate of the merit function at time step t;
definition of
Wherein θ represents a parameter of the agent policy;
is a jackpot; pi (a | s) is a strategy function; />
Is the state transition probability; gamma is a reward attenuation factor; π θ (at/st) represents the expectation of a jackpot for selecting an action value at a given policy π, state value st;
computing policy updates
After K time steps, updating by using small-batch SGD, and continuously iterating the steps until the model converges to obtain the trained intelligent agent pi * (as). Wherein, the first and the second end of the pipe are connected with each other,
is obtained by the following formula (I) in which,
wherein E represents expectation;
is the new and old strategy proportion, wherein theta is a parameter of the strategy neural network, pi θ (a t |s t ) Strategy representing time t, <' >>
The strategy of the previous iteration is adopted;
clip(r t (theta), 1-epsilon, 1+ epsilon) indicates that the proportion of the new strategy and the old strategy is controlled to be close to 1;
ε represents a hyperparameter.
And S4, inputting the state data of the person to be detected after executing the cardiovascular disease risk control scheme into the cardiovascular disease risk control intelligent body, obtaining the cardiovascular disease risk assessment value of the person to be detected in real time, and providing a further cardiovascular disease risk control scheme.
In summary, the implementation process of the cardiovascular disease risk control method is to input the current state of an individual to be detected, including directly changeable characteristics (DV), such as lifestyle characteristics; characteristics that cannot be changed (UV), such as age sex; characteristics (IV) that can be subject to indirect alteration, such as BMI index, blood glucose concentration, and the like.
Second, by cardiovascular disease risk prediction model f (x) V ) Judging the input cardiovascular disease risk of the individual to be detected; wherein, if the risk value of the cardiovascular disease risk of the individual to be detected is greater than 0.5, the trained cardiovascular disease risk control agent is used for adjusting the life habit (DV) characteristics. The agent enters his current state s 0 Outputting an optimal motion a in the motion space 0 I.e. to adjust a certain one of the lifestyle (DV) characteristicsValues, for example: the smoking amount is reduced by 1. At this time, the cardiovascular disease risk control intelligent body state is changed into s 1 。
Then, the cardiovascular disease risk prediction model f is recalled(s) 1 ) Judging whether the risk value of the cardiovascular disease risk of the individual to be detected is higher than 0.5, if so, continuing to output the action a 1 … … until the state changes to some s t The risk is below 0.5. In the process, the combination of actions performed by the agent { a } 0 ，a 1 ，a 2 ，...，a t-1 I.e. the individual's cardiovascular disease risk control program (i.e. lifestyle modification method). The individual to be tested can gradually adjust the living habits according to the cardiovascular disease risk control scheme, thereby realizing the reversal of the cardiovascular disease risk.
FIGS. 2 to 4 are graphs showing the effect of the cardiovascular disease risk prediction method according to the embodiment of the present invention; FIG. 2 is a diagram illustrating the convergence effect of the cardiovascular disease risk prediction agent provided by the embodiment of the present invention; FIG. 3 is a comparative graph of cardiovascular disease risk reduction before and after lifestyle modification by a cardiovascular disease risk prediction method according to an embodiment of the present invention; fig. 4 is a graph showing the effect of reducing the risk of cardiovascular diseases after 5-step lifestyle modification by the cardiovascular disease risk prediction method according to the embodiment of the present invention.
As can be seen from fig. 2, the cardiovascular disease risk prediction agent provided by the embodiment of the algorithm of the present invention has a very good convergence effect, and the cardiovascular disease risk prediction agent can have a stable high score reward value after 20000 training steps.
As can be seen from fig. 3, 500 high-risk (risk value over 0.5) individuals randomly sampled in the test set had an average risk value of 0.79 before lifestyle modification by the cardiovascular disease risk prediction agent; after being adjusted by the cardiovascular disease risk prediction agent, the average risk is reduced by about 50 percent, and is only 0.43. It is demonstrated that the method for predicting cardiovascular disease risk of the present invention can play a role in guiding lifestyle modification and reducing cardiovascular disease risk in general.
As can be seen from fig. 4, some of the sampled populations had a cardiovascular risk reduction of less than 0.5 after five steps of cardiovascular disease risk prediction agent adjustment. It can be seen that after each step of adjustment, the risk value gradually decreases. It is demonstrated that a cardiovascular disease risk prediction intelligence trained in accordance with the present invention can guide an individual in a reduced risk direction to make lifestyle adjustments.
As shown in fig. 5, the present invention provides a cardiovascular disease risk control system 500, which can be installed in an electronic device. Depending on the implemented functionality, the cardiovascular disease risk control system 500 may comprise the behavioral data acquisition unit 510, the status data screening unit 520 of a person at high risk of cardiovascular disease, the cardiovascular disease risk control scenario acquisition unit 530 and the cardiovascular disease risk control scenario adjustment unit 540. The module of the present invention, which may also be referred to as a unit, refers to a series of computer program segments that can be executed by a processor of an electronic device and can perform a fixed function, and are stored in a memory of the electronic device.
In the present embodiment, the functions regarding the respective modules/units are as follows:
a behavior data acquiring unit 510, configured to acquire behavior data of a person to be detected, where the behavior data is state data of an individual at a certain time;
a state data screening unit 520 of the cardiovascular disease high risk person, configured to input the behavior data into a pre-trained cardiovascular disease risk prediction model, and obtain a cardiovascular disease risk value of the person to be detected; screening the behavior data of the person to be detected with the cardiovascular disease risk value larger than a set risk threshold value as the state data of the person with high risk of cardiovascular disease;
a cardiovascular disease risk control scheme obtaining unit 530, configured to input the state data into a pre-trained cardiovascular disease risk control agent to obtain a cardiovascular disease risk control scheme; wherein the cardiovascular disease risk control agent is obtained based on reinforcement learning training;
the cardiovascular disease risk control scheme adjusting unit 540 is configured to input the state data of the person to be detected after executing the cardiovascular disease risk control scheme into the cardiovascular disease risk control agent, and obtain a cardiovascular disease risk assessment value of the person to be detected in real time; the cardiovascular disease risk control agent determines a further cardiovascular disease risk control regimen based on the cardiovascular disease risk assessment value.
The cardiovascular disease risk control system 500 of the invention can warn the cardiovascular disease risk of a person to be detected, and can help the person with high risk of cardiovascular disease to continuously adjust the living habits of the person by giving a cardiovascular disease risk control scheme, thereby achieving the technical effect of reducing the risk of future cardiovascular disease.
As shown in fig. 6, the present invention provides an electronic device 6 for a cardiovascular disease risk control method.
The electronic device 6 may comprise a processor 60, a memory 61 and a bus, and may further comprise a computer program, such as a cardiovascular disease risk control program 62, stored in the memory 61 and executable on said processor 60.
The memory 61 includes at least one type of readable storage medium, which includes flash memory, removable hard disk, multimedia card, card type memory (e.g., SD or DX memory, etc.), magnetic memory, magnetic disk, optical disk, etc. The memory 61 may in some embodiments be an internal storage unit of the electronic device 3, e.g. a removable hard disk of the electronic device 6. The memory 31 may also be an external storage device of the electronic device 6 in other embodiments, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like, provided on the electronic device 3. Further, the memory 61 may also include both an internal storage unit and an external storage device of the electronic device 6. The memory 61 may be used not only to store application software installed in the electronic device 6 and various types of data, such as codes of cardiovascular disease risk control programs, etc., but also to temporarily store data that has been output or is to be output.
The processor 60 may be formed of an integrated circuit in some embodiments, for example, a single packaged integrated circuit, or may be formed of a plurality of integrated circuits packaged with the same or different functions, including one or more Central Processing Units (CPUs), microprocessors, digital Processing chips, graphics processors, and combinations of various control chips. The processor 30 is a Control Unit (Control Unit) of the electronic device, connects various components of the whole electronic device by using various interfaces and lines, and executes various functions of the electronic device 6 and processes data by running or executing programs or modules (e.g., cardiovascular disease risk Control programs, etc.) stored in the memory 61 and calling data stored in the memory 61.
The bus may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. The bus is arranged to enable connection communication between the memory 61 and at least one processor 60 or the like.
Fig. 6 only shows an electronic device with components, and it will be understood by those skilled in the art that the structure shown in fig. 6 does not constitute a limitation of the electronic device 6, and may comprise fewer or more components than shown, or some components may be combined, or a different arrangement of components.
For example, although not shown, the electronic device 6 may further include a power supply (such as a battery) for supplying power to each component, and preferably, the power supply may be logically connected to the at least one processor 60 through a power management device, so that functions such as charge management, discharge management, and power consumption management are implemented through the power management device. The power supply may also include any component of one or more dc or ac power sources, recharging devices, power failure detection circuitry, power converters or inverters, power status indicators, and the like. The electronic device 6 may further include various sensors, a bluetooth module, a Wi-Fi module, and the like, which are not described herein again.
Further, the electronic device 6 may further include a network interface, and optionally, the network interface may include a wired interface and/or a wireless interface (such as a WI-FI interface, a bluetooth interface, etc.), which are generally used for establishing a communication connection between the electronic device 3 and other electronic devices.
Optionally, the electronic device 6 may further comprise a user interface, which may be a Display (Display), an input unit (such as a Keyboard), and optionally a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, or the like. The display, which may also be referred to as a display screen or display unit, is suitable for displaying information processed in the electronic device 6 and for displaying a visualized user interface.
It is to be understood that the embodiments described are illustrative only and are not to be construed as limiting the scope of the claims.
The cardiovascular disease risk control program 62 stored by the memory 61 in the electronic device 6 is a combination of instructions that, when executed in the processor 60, enable: acquiring behavior data of a person to be detected, wherein the behavior data is individual state data at a certain time; inputting the behavior data of the person to be detected into a pre-trained cardiovascular disease risk prediction model to obtain a cardiovascular disease risk value of the person to be detected; screening the behavior data of the person to be detected with the cardiovascular disease risk value larger than a set risk threshold value as the state data of the person with high risk of cardiovascular disease; inputting the state data into a pre-trained cardiovascular disease risk control intelligent agent to obtain a cardiovascular disease risk control scheme; wherein the cardiovascular disease risk control agent is obtained based on reinforcement learning training; and inputting the state data of the person to be detected after executing the cardiovascular disease risk control scheme into the cardiovascular disease risk control intelligent agent, obtaining the cardiovascular disease risk assessment value of the person to be detected in real time and providing a further cardiovascular disease risk control scheme.
Specifically, the specific implementation method of the processor 60 for the above instruction may refer to the description of the relevant steps in the corresponding embodiment of fig. 1, which is not repeated herein. It is emphasized that, to further ensure the privacy and security of the cardiovascular disease risk control procedure, the database may be stored with process data in the nodes of the blockchain in which the server cluster is located.
Further, the integrated modules/units of the electronic device 6, if implemented in the form of software functional units and sold or used as separate products, may be stored in a computer readable storage medium. The computer-readable medium may include: any entity or device capable of carrying said computer program code, a recording medium, a usb-disk, a removable hard disk, a magnetic diskette, an optical disk, a computer Memory, a Read-Only Memory (ROM).
An embodiment of the present invention further provides a computer-readable storage medium, where the storage medium may be nonvolatile or volatile, and the storage medium stores a computer program, and when the computer program is executed by a processor, the computer program implements: acquiring behavior data of a person to be detected, wherein the behavior data is individual state data at a certain time; inputting the behavior data of the person to be detected into a pre-trained cardiovascular disease risk prediction model to obtain a cardiovascular disease risk value of the person to be detected; screening the behavior data of the person to be detected with the cardiovascular disease risk value larger than a set risk threshold value as the state data of the person with high risk of cardiovascular disease; inputting the state data into a pre-trained cardiovascular disease risk control intelligent agent to obtain a cardiovascular disease risk control scheme; wherein the cardiovascular disease risk control agent is obtained based on reinforcement learning training; and inputting the state data of the person to be detected after executing the cardiovascular disease risk control scheme into the cardiovascular disease risk control intelligent agent, obtaining the cardiovascular disease risk assessment value of the person to be detected in real time and providing a further cardiovascular disease risk control scheme.
Specifically, the detailed implementation method of the computer program when being executed by the processor may refer to the description of the relevant steps in the cardiovascular disease risk control method in the embodiment, which is not repeated herein.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus, device and method can be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is only one logical functional division, and other divisions may be realized in practice.
The modules described as separate parts may or may not be physically separate, and parts displayed as modules may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, functional modules in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional module.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof.
The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference signs in the claims shall not be construed as limiting the claim concerned.
The block chain is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, a consensus mechanism, an encryption algorithm and the like. A block chain (Blockchain), which is essentially a decentralized database, is a series of data blocks associated by using a cryptographic method, and each data block contains information of a batch of network transactions, so as to verify the validity (anti-counterfeiting) of the information and generate a next block. The blockchain may include a blockchain underlying platform, a platform product service layer, an application service layer, and the like.
Furthermore, it is obvious that the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural. A plurality of units or means recited in the system claims may also be implemented by one unit or means in software or hardware. The terms second, etc. are used to denote names, but not any particular order.
Finally, it should be noted that the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting, and although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.
Claims (8)
1. A cardiovascular disease risk control system for use with a cardiovascular disease risk control method, the method comprising:
acquiring behavior data of a person to be detected, wherein the behavior data is state data of an individual at a certain time;
inputting the behavior data into a pre-trained cardiovascular disease risk prediction model to obtain a cardiovascular disease risk value of a person to be detected; screening the behavior data of the person to be detected with the cardiovascular disease risk value larger than a set risk threshold value as the state data of the person with high risk of cardiovascular disease;
the cardiovascular disease risk prediction model is as follows:
wherein, the first and the second end of the pipe are connected with each other,
in order to test the sample, the sample is,
is a cardiovascular disease risk prediction model
To pair
A desired prediction output of; Dis a regression modelDThe input of (1) is DV data and UV data, and the output is IV data;
DV is a directly variable feature including daily smoking volume, daily drinking volume, and daily exercise volume;
UV is an unchangeable characteristic including age, gender, and territory;
IV is an indirectly variable feature including BMI index, blood glucose concentration, blood pressure, body fat rate and waist circumference;
inputting the state data into a pre-trained cardiovascular disease risk control intelligent agent to obtain a cardiovascular disease risk control scheme; wherein the cardiovascular disease risk control agent is obtained based on reinforcement learning training;
inputting the state data of the person to be detected after executing the cardiovascular disease risk control scheme into a cardiovascular disease risk control intelligent body, and obtaining a cardiovascular disease risk assessment value of the person to be detected in real time;
the cardiovascular disease risk control agent determines a further cardiovascular disease risk control regimen based on the cardiovascular disease risk assessment value.
2. The cardiovascular disease risk control system of claim 1, wherein the reinforcement learning-based training method for the cardiovascular disease risk control agent comprises:
building an intelligent environment for cardiovascular disease risk control and a reinforcement learning model for cardiovascular disease risk control; wherein the intelligent environment is
(ii) a Wherein the content of the first and second substances,Sis a state space comprising cardiovascular disease features; the cardiovascular disease trait includes a directly alterable trait, an unalterable trait, and an indirectly alterable trait;Afor a motion space comprising directly changeable features,Ris a function of the reward, which is,
in order to be a function of the probability of state transitions,
is a reward attenuation factor;
evaluating the sequence action output by the reinforcement learning model for cardiovascular disease risk control by using the intelligent agent environment, giving a reward corresponding to the sequence action and the sequence action at the next moment, and carrying out centralized training with the maximization of accumulated reward as a target to obtain the intelligent agent
(ii) a Wherein the agent
The input of (1) is the state of the environment, and the output is the sequence action selected to be executed under the state of the environment;
and iterating to the convergence of the intelligent agent model through a generalized dominance function and a strategy to obtain the trained cardiovascular disease risk control intelligent agent based on reinforcement learning.
3. The cardiovascular disease risk control system of claim 2, wherein the cardiovascular disease risk prediction model is obtained by optimal feature variable set training; the method for acquiring the optimal characteristic variable set comprises the following steps:
inputting a data set containing cardiovascular disease behavior characteristics as a training set into a behavior characteristic importance evaluation model;
performing importance evaluation on the behavior characteristics of the cardiovascular diseases in the training set through the behavior characteristic importance evaluation model, and outputting the behavior characteristics of the cardiovascular diseases with the importance evaluation meeting a set threshold value as an influence factor set;
and taking the influence factor set as an optimal characteristic variable set.
4. The cardiovascular disease risk control system of claim 2, wherein the jackpot is derived by the equation:
wherein the content of the first and second substances,
the score is given as the background score,
，
in order to reward the scaling factor(s),ca coefficient representing the degree of tolerance to the action,
，
is in a state ofS t Is operated asa t The output value of the cardiovascular disease risk prediction model;
in order to penalize the score parameter,
is a final outcome parameter, and when the agent wins
When the agent fails
If the agent is in a non-terminal state, then
。
5. The cardiovascular disease risk control system of claim 2,
the method for converging the intelligent agent model through the generalized dominance function and the strategy iteration comprises the following steps:
estimation with merit function
And the number of the first and second electrodes,
；
wherein the content of the first and second substances,
is an estimate of the merit function at time step t,
in order to be a policy-raising function,
evaluating a function for the policy;
computing policy updates
And pass throughKPerforming strategy updating by using small-batch SGD (serving gateway device) in each time step; wherein the content of the first and second substances,
a parameter representing an agent policy;
is a jackpot;
is a policy function;
is the state transition probability;
a reward attenuation factor;
is a loss function;
and continuously iterating the steps until the intelligent agent model converges.
6. The cardiovascular disease risk control system of claim 5, wherein the loss function
Is obtained by the following formula (I) in which,
wherein, the first and the second end of the pipe are connected with each other,Eindicates a desire;
is the new and old policy proportion, wherein
In order to be a parameter of the neural network of the strategy,
the policy representing the time t is that of the time,
is the strategy of the previous iteration;
7. An electronic device, characterized in that the electronic device comprises:
at least one processor; and (c) a second step of,
a memory communicatively coupled to the at least one processor; wherein, the first and the second end of the pipe are connected with each other,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the steps in the cardiovascular disease risk control system of any of claims 1-6.
8. A computer-readable storage medium, storing a computer program, wherein the computer program, when executed by a processor, implements a cardiovascular disease risk control system as claimed in any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111339208.2A CN113940640B (en) | 2021-11-12 | 2021-11-12 | Cardiovascular disease risk control method, system and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111339208.2A CN113940640B (en) | 2021-11-12 | 2021-11-12 | Cardiovascular disease risk control method, system and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113940640A CN113940640A (en) | 2022-01-18 |
| CN113940640B true CN113940640B (en) | 2023-04-18 |
Family
ID=79338011
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111339208.2A Active CN113940640B (en) | 2021-11-12 | 2021-11-12 | Cardiovascular disease risk control method, system and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113940640B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115223721B (en) * | 2022-07-12 | 2023-05-16 | 曜立科技(北京)有限公司 | System for analyzing similarity of cardiovascular and cerebrovascular disease tube numbers of patient based on metadata |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109087708A (en) * | 2018-09-20 | 2018-12-25 | 深圳先进技术研究院 | Model training method, device, equipment and storage medium for patch segmentation |
| CN110693489A (en) * | 2019-10-28 | 2020-01-17 | 杭州电子科技大学 | Myocardial infarction detection method based on multi-classifier reinforcement learning |
Family Cites Families (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10265028B2 (en) * | 2012-08-16 | 2019-04-23 | Ginger.io, Inc. | Method and system for modeling behavior and heart disease state |
| US20170329905A1 (en) * | 2016-05-12 | 2017-11-16 | Siemens Healthcare Gmbh | Life-Long Physiology Model for the Holistic Management of Health of Individuals |
| JP6881289B2 (en) * | 2017-12-27 | 2021-06-02 | オムロンヘルスケア株式会社 | Disease risk predictors, methods and programs |
| US20190295727A1 (en) * | 2018-03-23 | 2019-09-26 | American Heart Association, Inc. | System and Method for Assessing Heart Health and Communicating the Assessment to a Patient |
| US10811139B1 (en) * | 2018-06-13 | 2020-10-20 | Clarify Health Solutions, Inc. | Computer network architecture with machine learning and artificial intelligence and dynamic patient guidance |
| US20200111574A1 (en) * | 2018-10-03 | 2020-04-09 | The University Of Trinidad And Tobago | System and method for deriving personalized cardiovascular disease risk assessments |
| US20200373017A1 (en) * | 2019-05-24 | 2020-11-26 | The Trustees Of Princeton University | System and method for intelligence crowdsourcing reinforcement learning for clinical pathway optimization |
| US11883208B2 (en) * | 2019-08-06 | 2024-01-30 | Medtronic Minimed, Inc. | Machine learning-based system for estimating glucose values based on blood glucose measurements and contextual activity data |
| WO2021061702A1 (en) * | 2019-09-23 | 2021-04-01 | The University Of Chicago | Method of creating zero-burden digital biomarkers for disorders, and exploiting co-morbidity patterns to drive early intervention |
| US20210319900A1 (en) * | 2020-04-14 | 2021-10-14 | Siemens Healthcare Gmbh | Cardiovascular assessment of patients suspected of having covid-19 |
| CN112133445A (en) * | 2020-10-21 | 2020-12-25 | 万达信息股份有限公司 | Cardiovascular disease management service method and system |
-
2021
- 2021-11-12 CN CN202111339208.2A patent/CN113940640B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109087708A (en) * | 2018-09-20 | 2018-12-25 | 深圳先进技术研究院 | Model training method, device, equipment and storage medium for patch segmentation |
| CN110693489A (en) * | 2019-10-28 | 2020-01-17 | 杭州电子科技大学 | Myocardial infarction detection method based on multi-classifier reinforcement learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113940640A (en) | 2022-01-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kim et al. | A deep learning model for real-time mortality prediction in critically ill children | |
| US9370689B2 (en) | System and methods for providing dynamic integrated wellness assessment | |
| US11817218B2 (en) | Physiologic severity of illness score for acute care patients | |
| WO2014033681A2 (en) | Modeling techniques for predicting mortality in intensive care units | |
| CN116453706B (en) | Hemodialysis scheme making method and system based on reinforcement learning | |
| US20200074573A1 (en) | System and method for providing a patient-specific prediction model in a user application for effectiveness determinations | |
| CN107249434B (en) | Robust classifier | |
| Zhu et al. | Blood glucose prediction in type 1 diabetes using deep learning on the edge | |
| CN110046757B (en) | Outpatient clinic volume prediction system and prediction method based on LightGBM algorithm | |
| KR20180099193A (en) | Method, server and system for providing disease predictions service usiing multiple disease prediction models | |
| US20210358628A1 (en) | Digital companion for healthcare | |
| CN113940640B (en) | Cardiovascular disease risk control method, system and storage medium | |
| US20220059226A1 (en) | Device and Method for Inferring Depressive State and Program for Same | |
| KR20180099185A (en) | Method, server and system for generating disease prediction models | |
| JP7266548B2 (en) | Apparatus, program and method for estimating physical condition using training stress information | |
| CN112542242A (en) | Data transformation/symptom scoring | |
| CN112115322A (en) | User grouping method and device, electronic equipment and storage medium | |
| KR102298119B1 (en) | Method and computer device for providing analytical information related to sleep | |
| US20220406466A1 (en) | A method for determining a risk score for a patient | |
| Wang et al. | A high-fidelity model to predict length of stay in the neonatal intensive care unit | |
| CN111899862B (en) | Breathing machine parameter acquisition method, device, equipment and storage medium | |
| Agrawal et al. | Lung transplant outcome prediction using unos data | |
| JP2024518693A (en) | Medical Event Prediction Using Personalized Dual-Channel Coupling Networks | |
| AU2021363110A1 (en) | Method and system for personalized prediction of infection and sepsis | |
| CN107799182A (en) | The evaluation method and electronic equipment of complication and complication factor of influence |
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 |