CN117976209A

CN117976209A - Medical event probability determination method, device, equipment and computer storage medium

Info

Publication number: CN117976209A
Application number: CN202410103884.7A
Authority: CN
Inventors: 李静; 张丽华; 王薇; 刘佳敏
Original assignee: Fuwai Hospital of CAMS and PUMC
Current assignee: Fuwai Hospital of CAMS and PUMC
Priority date: 2024-01-25
Filing date: 2024-01-25
Publication date: 2024-05-03

Abstract

The application discloses a medical event probability determination method, a medical event probability determination device, medical event probability determination equipment and a computer storage medium. The method comprises the following steps: acquiring a numerical value of each of a plurality of predictors of a target patient hospitalized for acute heart failure; searching the readmission risk score corresponding to the numerical value of each prediction index from the corresponding relation between a plurality of value intervals of each prediction index and the readmission risk score; adding and processing the readmission risk scores to obtain a readmission risk total score of the target patient; searching the target readmission probability corresponding to the readmission risk total score of the target patient from the corresponding relation between the readmission risk total score and the readmission occurrence probability, and obtaining the probability of readmission of the target patient. According to the embodiment of the application, the efficiency of determining the readmission probability of the target patient is improved, and the requirement of clinical diagnosis and treatment on personalized risk assessment is met.

Description

Medical event probability determination method, device, equipment and computer storage medium

Technical Field

The application belongs to the field of heart failure disease research, and particularly relates to a medical event probability determination method, a device, equipment and a computer storage medium.

Background

Heart Failure (HF) is a serious and end-stage manifestation of various heart diseases, and has the characteristics of high hospitalization rate, high death rate, etc., which brings serious burden to families and society.

The diagnosis and treatment work with the main purposes of improving the life quality of acute HF patients, improving prognosis and reducing disease burden is not satisfactory, and the method has higher requirements on developing the disease risk and prognosis related research of the acute HF patients.

The rate of readmission of acute HF patients 30 days after discharge is an important indicator for the evaluation of HF medical quality. The risk prediction model for readmission of acute HF patients 30 days after discharge has progressed in recent years. Currently, tens of acute HF patients are subjected to hospital readmission risk prediction model researches abroad, and the model researches are mainly carried out by European and American countries. However, most models have low prediction efficiency and accuracy, and cannot meet the requirement of clinical diagnosis and treatment on personalized risk assessment.

Disclosure of Invention

The embodiment of the application provides a method, a device, equipment and a computer storage medium for determining the probability of a medical event, which can improve the efficiency of determining the probability of readmission of a target patient and meet the requirement of clinical diagnosis and treatment on personalized risk assessment.

In a first aspect, an embodiment of the present application provides a method for determining a probability of a medical event, including:

Acquiring a numerical value of each of a plurality of predictors of a target patient hospitalized for acute heart failure;

searching the readmission risk score corresponding to the numerical value of each prediction index from the corresponding relation between a plurality of value intervals of each prediction index and the readmission risk score;

adding and processing the readmission risk scores to obtain a readmission risk total score of the target patient;

searching the target readmission probability corresponding to the readmission risk total score of the target patient from the corresponding relation between the readmission risk total score and the readmission occurrence probability, and obtaining the probability of readmission of the target patient.

In some embodiments, before searching for the readmission risk score corresponding to the value of each predictor from the correspondence between the multiple value intervals of each predictor and the readmission risk score, the method further includes:

Acquiring a value range of a first prediction index, a basic risk reference value of a first evaluation index and a constant value of a crowd hospitalized with acute heart failure; the first prediction index is any one of a plurality of prediction indexes, and the constant value is used for indicating the value corresponding to each 1 time-sharing increase of the risk score;

dividing the value range into a plurality of value intervals, and determining index reference values of each value interval;

calculating the distance between the index reference value and the basic risk reference value of each value interval;

calculating the ratio between the distance and the constant value to obtain a readmission risk score corresponding to each value interval of the first prediction index;

and obtaining the corresponding relation between the multiple value intervals of the first prediction index and the readmission risk score based on the value intervals of the first prediction index and the readmission risk scores corresponding to the value intervals.

In some embodiments, searching for a readmission risk score corresponding to the value of each predictor from the correspondence between the plurality of value intervals of each predictor and the readmission risk score includes:

Determining a target value range corresponding to the numerical value of each prediction index from the value ranges of a plurality of value intervals of each prediction index;

and searching risk scores corresponding to the target value ranges from the corresponding relations between the multiple value intervals of each prediction index and the readmission risk scores to obtain the risk scores corresponding to the numerical values of the prediction indexes.

In some embodiments, the method further includes, before searching for a target readmission probability corresponding to the target patient readmission risk total score from a correspondence between the readmission risk total score and the readmission probability, and obtaining the target patient readmission probability:

acquiring a total readmission risk score of a first patient hospitalized with acute heart failure, a numerical value of each of the P predictors, a regression coefficient of each predictor, an average numerical value of each of the P predictors of the patient hospitalized with acute heart failure, and an average readmission probability of the patient hospitalized with acute heart failure; the first patient is any patient hospitalized for acute heart failure; p is an integer greater than 1;

Determining a sum of a base risk value and an individual risk value of the first patient based on the numerical value of each predictor of the P predictors of the first patient and the regression coefficient of each predictor;

Determining an average risk of readmission of the population hospitalized for acute heart failure based on the average value of each predictor of the P predictors of the population hospitalized for acute heart failure and the regression coefficient of each predictor;

determining a readmission probability of the first patient based on the average readmission probability, a sum of the base risk value and the individual risk value of the first patient, and the average risk of readmission of the population;

And determining the corresponding relation between the total risk score of the readmission and the probability of readmission occurrence based on the total risk score of the first patient and the probability of readmission of the first patient.

In some embodiments, prior to obtaining the total readmission risk score for the first patient hospitalized for acute heart failure, the value of each predictor of the P predictors, the regression coefficient for each predictor, the average value of each predictor of the P predictors for the population hospitalized for acute heart failure, and the average probability of readmission, the method further comprises:

Acquiring a baseline risk function of a second patient hospitalized for acute heart failure at a target time, an interval time from discharge to readmission, and a numerical value of each of the P predictors; the second patient is any patient hospitalized for acute heart failure;

determining a risk function value for the second patient at the target time based on the interval duration from discharge to readmission;

The regression coefficient of each predictor is determined based on the risk function value of each second patient at the target time, the interval duration from discharge to readmission, the value of each predictor of the P predictors.

In some embodiments, the method further comprises:

Determining a risk index of readmission of the target patient when a preset time is spent after discharge of the target patient based on the probability of readmission of the target patient;

and determining the readmission risk level of the target patient based on the risk index of readmission when the target patient is discharged for a preset time period.

In a second aspect, an embodiment of the present application provides a medical event probability determining apparatus, including:

An acquisition module for acquiring a value of each of a plurality of predictors of a target patient hospitalized for acute heart failure;

the searching module is used for searching the readmission risk score corresponding to the numerical value of each prediction index from the corresponding relation between the multiple value intervals of each prediction index and the readmission risk score;

The adding module is used for adding the risk scores of the readmission to obtain a total risk score of the readmission of the target patient;

the searching module is further used for searching the target readmission probability corresponding to the readmission risk total score of the target patient from the corresponding relation between the readmission risk total score and the readmission occurrence probability, and obtaining the readmission probability of the target patient.

In a third aspect, an embodiment of the present application provides a medical event probability determining apparatus, the apparatus including: a processor and a memory storing computer program instructions;

The processor, when executing the computer program instructions, implements the medical event probability determination method according to any one of the first aspects.

In a fourth aspect, embodiments of the present application provide a computer readable storage medium having stored thereon computer program instructions which, when executed by a processor, implement a medical event probability determination method as in any of the first aspects.

In a fifth aspect, embodiments of the present application provide a computer program product, instructions in which, when executed by a processor of an electronic device, cause the electronic device to perform the medical event probability determination method according to any of the first aspects.

According to the medical event probability determination method, device and equipment and computer storage medium, the readmission risk score corresponding to the numerical value of each prediction index of the target patient can be searched from the corresponding relation between the multiple value intervals of each prediction index of the acute heart failure and the readmission risk score, the readmission risk score of the target patient is obtained by adding and processing the readmission risk scores, namely, the target patient readmission risk total score is comprehensively determined according to the multiple prediction indexes, the target readmission probability corresponding to the target patient readmission risk total score is searched from the corresponding relation between the readmission risk total score and the readmission occurrence probability of the target patient, the readmission probability of the target patient is obtained, the accuracy of determining the readmission probability of the target patient is effectively improved, in addition, the readmission probability of the target patient can be rapidly found through the corresponding relation between the multiple value intervals of each prediction index and the readmission risk score and the readmission occurrence probability, the clinical evaluation risk is improved, and the clinical evaluation requirement of the target patient readmission probability is met.

Drawings

In order to more clearly illustrate the technical solution of the embodiments of the present application, the drawings that are needed to be used in the embodiments of the present application will be briefly described, and it is possible for a person skilled in the art to obtain other drawings according to these drawings without inventive effort.

FIG. 1 is a flow chart of a method for determining probability of medical events according to one embodiment of the present application;

FIG. 2 is a flow chart of a medical event probability determination method provided by another embodiment of the present application;

FIG. 3 is a flow chart of a medical event probability determination method provided by yet another embodiment of the present application;

FIG. 4 is a schematic diagram of a medical event probability determination apparatus according to an embodiment of the present application;

fig. 5 is a schematic structural view of a medical event probability determining apparatus provided in an embodiment of the present application.

Detailed Description

Features and exemplary embodiments of various aspects of the present application will be described in detail below, and in order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be described in further detail below with reference to the accompanying drawings and the detailed embodiments. It should be understood that the particular embodiments described herein are meant to be illustrative of the application only and not limiting. It will be apparent to one skilled in the art that the present application may be practiced without some of these specific details. The following description of the embodiments is merely intended to provide a better understanding of the application by showing examples of the application.

It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, 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 … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

In addition, the technical scheme of the application can acquire, store, use, process and the like the data, which accords with the relevant regulations of national laws and regulations.

The rate of readmission of acute HF patients 30 days after discharge is an important indicator for the evaluation of HF medical quality. The risk prediction model for readmission of acute HF patients 30 days after discharge has progressed in recent years. Currently, tens of acute HF patients are subjected to hospital readmission risk prediction model researches abroad, and the model researches are mainly carried out by European and American countries. However, most models only have a medium-low level risk prediction capability, and cannot meet the requirement of clinical diagnosis and treatment on personalized risk assessment. In addition, the 30-day readmission risk prediction model for acute HF patients, which is internationally established based on western populations, is not applicable to national people.

Based on the above problems, the application provides a medical event probability determination method, device, equipment and computer storage medium, which can search the readmission risk score corresponding to the numerical value of each prediction index of a target patient from the corresponding relation between a plurality of value intervals of each prediction index of acute heart failure and the readmission risk score, add and process each readmission risk score to obtain the total readmission risk score of the target patient, that is, comprehensively determine the total readmission risk score of the target patient according to a plurality of prediction indexes, search the target readmission probability corresponding to the total readmission risk score of the target patient from the corresponding relation between the total readmission risk score of the target patient and the readmission probability, obtain the readmission probability of the target patient, effectively improve the accuracy of determining the readmission probability of the target patient, and further quickly search the readmission probability of the target patient through the corresponding relation between the plurality of value intervals of each prediction index and the readmission risk score and the readmission probability, and improve the clinical evaluation requirement of the determined individual clinical risk.

In order to solve the problems in the prior art, the embodiment of the application provides a medical event probability determination method, a device, equipment and a computer storage medium. The method for determining the probability of a medical event provided by the embodiment of the application is first described below.

Fig. 1 is a flow chart of a medical event probability determination method according to an embodiment of the present application. As shown in fig. 1, the medical event probability determining method provided by the embodiment of the application includes the following steps: S101-S104.

S101: a value for each of a plurality of predictors for a target patient hospitalized for acute heart failure is obtained.

In one example, the prediction index may be a clinical characteristic index, a laboratory examination index, a health condition index, or a hospitalization utilization index. Wherein, the clinical characteristic index comprises the following various indexes: systolic blood pressure (systolic blood pressure, SBP) levels, new york heart function classification (New York Heart Association Classification for Heart Failure, NYHA) for admission; the health condition index includes the following various indexes: an admitted N-terminal B-type natriuretic peptide precursor (N-terminal prohormone of brain natriuretic peptide, NT-proBNP) level, an admitted serum creatinine (Serum creatine, CRE) level, an admitted glycosylated Hemoglobin (glycated Hemoglobin A c, hbA1 c) level, an admitted Hemoglobin (Hemoglobin, hgb) level; the health condition indicators include a Kansas city myocardial disease questionnaire (KANSAS CITY Cardiomyopathy Questionnaire, KCCQ) score; the hospitalization service utilization condition index comprises the following various indexes: length of hospitalization, discharge symptoms.

In one embodiment, a value for each predictor of a plurality of predictors for a target patient hospitalized for acute heart failure may be obtained.

S102: searching the readmission risk score corresponding to the numerical value of each prediction index from the corresponding relation between a plurality of value intervals of each prediction index and the readmission risk score.

It can be understood that the values of the predictors can evaluate the conditions of different aspects of the target patient, so that the values of the predictors have an association relationship with the risk degree of readmission of the target patient, that is, a correspondence exists between a plurality of value intervals of each predictor and the readmission risk score.

Therefore, the correspondence between the multiple value intervals of each predictor and the readmission risk score may be predetermined, and the readmission risk score corresponding to the numerical value of each predictor may be searched from the correspondence between the multiple value intervals of each predictor and the readmission risk score.

In some embodiments of the present application, index Data of a patient hospitalized for acute heart failure may be obtained, regression processing is performed on the index Data using a minimum Absolute contraction and selection algorithm (LASSO) to reduce complexity of the index Data, the regression processed index Data is obtained, a random survival forest model (Random Survival Forest, RSF) is used to randomly extract samples from the raw Data in a form of put-back to create a plurality of sample subsets, and 37% of Data in each sample subset is excluded as Out-of-Bag Data (OOB). And randomly selecting characteristics for each sample to construct a corresponding binary survival tree. P candidate variables are randomly selected on each node of the tree, the nodes are split by using the candidate variables which maximize survival differences among the child nodes, the tree is grown to the complete size under the constraint of a readmission event of the leaf node, a cumulative risk function (cumulative hazard function, CHF) is calculated on each tree, the average value of the integrated cumulative risk function is obtained, and the prediction error of the integrated cumulative risk function is calculated by using OOB.

In one embodiment, a model may be constructed by choosing a number of features of the input data as its split nodes by the RSF model, and evaluating the variable importance by a variable importance method (variable importance, VIMP). The principle of the VIMP method is that OOB is put into a survival tree, and is randomly distributed to any child node; calculating a new total accumulated risk; VIMP is the difference between the original error rate and the new error rate. Thus, a larger VIMP means that the larger the influence of this variable on the accuracy of the model, the higher the importance of this variable. Through variable screening, a prediction index with larger VIMP can be screened out according to the function of each variable in the process of establishing a model.

In some embodiments of the application, the readmission risk score may be a readmission risk score within 30 days of discharge of the target patient.

In some embodiments of the present application, S102 may include the steps of:

In one embodiment, as shown in table 1, the value ranges of the multiple value intervals of each predictor may be predetermined, and the readmission risk score corresponding to each value range of each predictor may be predetermined. Under the condition that the readmission probability needs to be determined for the target patient, a target value range corresponding to the numerical value of each prediction index can be determined according to the numerical value of each prediction index in the multiple prediction indexes of the target patient.

Table 1 inpatient readmission risk score for acute heart failure

S103: and adding the readmission risk scores to obtain the total readmission risk score of the target patient.

In one embodiment, the readmission risk scores may be summed up to comprehensively evaluate the readmission risk of the target patient and obtain a total readmission risk score for the target patient.

S104: searching the target readmission probability corresponding to the readmission risk total score of the target patient from the corresponding relation between the readmission risk total score and the readmission occurrence probability, and obtaining the probability of readmission of the target patient.

It may be appreciated that the total readmission risk score and the readmission probability have an association relationship, a correspondence between the readmission probability and the readmission risk score of different readmission risks may be predetermined, and after the total readmission risk score of the target patient is obtained, the probability of readmission of the target patient may be obtained by searching for the target readmission probability corresponding to the total readmission risk score of the target patient from the correspondence between the total readmission risk score and the readmission probability.

For example, the correspondence between the total readmission risk score and the probability of readmission occurrence may be as shown in table 2:

TABLE 2 correspondence between readmission risk scores and readmission occurrence probabilities for patients hospitalized for acute heart failure

According to the medical event probability determination method provided by the embodiment of the application, the readmission risk scores corresponding to the numerical values of each prediction index of the target patient are searched from the corresponding relations between the multiple value intervals of each prediction index of the acute heart failure and the readmission risk scores, and the readmission risk scores of the target patient are subjected to addition processing, so that the readmission risk total scores of the target patient are obtained, namely, the target patient readmission risk total scores are comprehensively determined according to the multiple prediction indexes, the target readmission probability corresponding to the target patient readmission risk total scores is searched from the corresponding relations between the readmission risk total scores and the readmission occurrence probability of the target patient, so that the readmission probability of the target patient is obtained, the accuracy of determining the readmission probability of the target patient is effectively improved, and in addition, the readmission probability of the target patient can be rapidly searched through the corresponding relations between the multiple value intervals of each prediction index and the readmission risk score, and the clinical evaluation risk of individual requirements are met.

In some embodiments of the present application, as shown in fig. 2, before S102, the method may further include: S201-S205.

S201: and acquiring a value range of a first prediction index, a basic risk reference value of a first evaluation index and a constant value of the patient hospitalized by the acute heart failure.

In the embodiment of the present application, the first prediction index is any one of a plurality of prediction indexes, and the constant value is used to indicate a value corresponding to each 1 time-sharing increase of the risk score.

The above-mentioned value ranges may be preset according to actual requirements, and basic risk reference values and constant values may be configured for each value range in advance. Each 1 score increase in the indicated risk score corresponds to a constant value.

In one embodiment, a range of values for a first predictor of a patient hospitalized for acute heart failure, a base risk reference value for a first evaluation indicator, and a constant value may be obtained.

S202: dividing the value range into a plurality of value intervals, and determining index reference values of each value interval.

In the embodiment of the present application, the above-mentioned value range may be divided into a plurality of value intervals, and the index reference value of each value interval is determined respectively. The index reference value may be preset according to the actual requirement.

S203: and calculating the distance between the index reference value and the basic risk reference value of each value interval.

In one example, the distance between the index reference value and the base risk reference value of the value interval may be calculated by the following formula (1):

D＝(W_ij-W_iREF)×β_i (1)

Wherein D is the distance, W _ij is the index reference value corresponding to the ith value interval and the jth value interval, W _iREF is the basic risk reference value, and beta _i is the regression coefficient corresponding to the ith prediction index.

S204: and calculating the ratio between the distance and the constant value to obtain the readmission risk score corresponding to each value interval of the first prediction index.

In the embodiment of the application, the readmission risk score corresponding to the value interval can be obtained through calculation according to the following formula (2):

Wherein Points _ij is a readmission risk score and B is a constant value.

S205: and obtaining the corresponding relation between the multiple value intervals of the first prediction index and the readmission risk score based on the value intervals of the first prediction index and the readmission risk scores corresponding to the value intervals.

In one embodiment, after obtaining the readmission risk scores corresponding to the value intervals of the first predictor, the correspondence between the value intervals of the first predictor and the readmission risk scores may be obtained.

In some embodiments of the present application, there may be a plurality of first predictors, so that a correspondence between a plurality of value intervals of each first predictor and a readmission risk score may be obtained.

Through the corresponding relation between the multiple value intervals of the first prediction index and the readmission risk score, after the numerical value of each prediction index in the multiple prediction indexes of the target patient is obtained, the numerical value corresponding to each prediction index of the target patient can be quickly found, so that the efficiency of evaluating the readmission risk of the target patient can be improved.

In some embodiments of the present application, as shown in fig. 3, the method may further include the following steps before S104: S301-S305.

S301: the total readmission risk score of the first patient hospitalized for acute heart failure, the value of each predictor of the p predictors, the regression coefficient of each predictor, the average value of each predictor of the p predictors of the patient hospitalized for acute heart failure, and the average readmission probability are obtained.

In the embodiment of the present application, the first patient may be any patient hospitalized for acute heart failure.

In the embodiment of the application, p is an integer greater than 1.

In one embodiment, the total readmission risk score for the first patient hospitalized for acute heart failure, the value of each predictor of the p predictors, the regression coefficient of each predictor, the average value of each predictor of the p predictors for the patient hospitalized for acute heart failure, and the average readmission probability may be obtained in advance.

In the embodiment of the present application, the regression coefficient is used to represent the influence degree of the corresponding prediction index on the readmission risk of the first patient.

S302: the sum of the base risk value and the individual risk value of the first patient is determined based on the value of each predictor of the p predictors of the first patient, and the regression coefficient of each predictor.

In one embodiment, the sum of the base risk value and the individual risk value for the first patient may be calculated by the following equation (3):

Wherein A is the sum of the basic risk value and the individual risk value of the first patient, p is the number of predictors of the first patient, beta _j is the regression coefficient of the ith predictor, and X _i is the value of the ith predictor.

In one example, equation (3) may be calculated by:

k ₁、K₂ is a coefficient.

S303: based on the average value of each predictor of the p predictors of patients hospitalized for acute heart failure and the regression coefficients K ₁ and K ₂ of each predictor, the average risk of readmission of patients hospitalized for acute heart failure is determined.

In one embodiment, the average risk of readmission of a patient hospitalized for acute heart failure may be determined by the following equation (4):

wherein B is the average risk of readmission of patients hospitalized for acute heart failure, The average value of the ith predictive index of the patient hospitalized for acute heart failure, p is the number of predictive indexes.

For example, equation (4) may be calculated by K ₃-K₁₉ as the regression coefficient of each predictor:

wherein, K ₁ to K ₁₆ are all coefficients.

S304: the readmission probability of the first patient is determined based on the average readmission probability, the sum of the base risk value and the individual risk value of the first patient, and the average risk of readmission of the population.

In one embodiment, the readmission probability of the first patient may be determined by the following equation (6):

Wherein, The readmission probability of the first patient is S ₀ (t), which is the average readmission probability.

It should be noted that, the survival analysis may be expressed by a survival function, which is called cumulative survival rate, and the formula is: s (T) =p (T > T), where T is the time-to-live, the meaning of the function is the probability that the time-to-live is greater than the time point T. When t=0, S (T) =1, S (T) does not increase with increasing T, and 1-S (T) is a cumulative distribution function representing the probability that the survival time T does not exceed T.

S305: and determining the corresponding relation between the total risk score of the readmission and the probability of readmission occurrence based on the total risk score of the first patient and the probability of readmission of the first patient.

In one embodiment, since there is a one-to-one correspondence between the total risk score of the first patient and the readmission probability of the first patient, the correspondence between the total readmission risk score and the readmission probability of the first patient may be determined based on the total risk score of the first patient and the readmission probability of the first patient.

It can be understood that, because the first patient can be any patient in hospital due to acute heart failure, under the condition that a plurality of first patients exist, the corresponding relation between the total readmission risk scores and the readmission occurrence probability of the plurality of first patients can be obtained, and under the condition that the readmission probability of the target patient is determined, the readmission occurrence probability of the target patient can be quickly found out from the corresponding relation between the total readmission risk scores and the readmission occurrence probability of the plurality of target patients, thereby effectively improving the efficiency of determining the readmission occurrence probability of the target patient.

In some embodiments of the present application, prior to S301, the method may further comprise the steps of:

Acquiring a baseline risk function h ₀ (t) of a second patient hospitalized for acute heart failure at a target time, an interval time t from discharge to readmission, and a numerical value x of each predictive index of the P predictive indexes; the second patient is any patient hospitalized for acute heart failure;

Determining a risk function value h (t, x) of the second patient at the target time based on the interval duration from discharge to readmission;

The regression coefficient of each predictor is determined based on the risk function value h (t, x) of each second patient at the target time, the interval time period t from discharge to readmission, and the value x of each predictor of the P predictors.

In one embodiment, the risk function value may be calculated by the following equation (7) and equation (8):

h(t，x)＝h₀(t)exp(β₁x₁+β₂x₂+……+β_px_p) (7)

Wherein h (T, x) is a risk function value of the second patient at the target time, h ₀ (T) is a baseline risk function of the second patient at the target time, T is an interval time from discharge to readmission, P is the number of predictors, T is an interval time from discharge to the current time, beta ₁ to beta _p are regression coefficients respectively corresponding to P predictors, x ₁ to x _p are values of P predictors, P (T < t+Δt|t > T, x) is a probability, and formula (8) represents a limit value of a ratio of probability of readmission to Δt in a short time from T to Δt after the second patient with x is not readmitted at the T time.

In one embodiment, the risk function value h (t, x) corresponding to each second patient, the interval time period t from discharge to readmission, and the numerical value x of each of the P predictors are substituted into the above formula (7) and formula (8), so as to obtain an equation including P unknowns β ₁ to β _p, and the P equations are subjected to fitting processing, so as to obtain respective values of β ₁ to β _p, that is, determine the regression coefficient of each predictor.

In some embodiments of the present application, the method may further comprise the steps of:

In one embodiment, the index value of the risk index of readmission of the target patient at the preset time after discharge of the target patient may be a score obtained by scoring the target patient according to the probability of readmission, and the readmission risk level of the target patient may be determined according to the score, so that the readmission possibility of the target patient may be more intuitively determined.

For example, patients with a probability of readmission of the target patient in the interval 0-25%, i.e. patients scored between 0-13, may be determined to be at low risk; the risk level of readmission of the target patient, namely the patient with the probability of readmission of the target patient within the range of 26% -75%, is determined to be medium risk; patients with a probability of readmission of the target patient in the interval 76% -100%, i.e. patients with a score above 18, are determined to be at high risk.

Based on the medical event probability determining method provided by the above embodiment, a specific implementation of the medical event probability determining apparatus provided by the embodiment of the present application will be described with reference to fig. 4.

Referring to fig. 4, a schematic structural diagram of a medical event probability determining apparatus according to an embodiment of the present application, the medical event probability determining apparatus 400 includes:

An acquisition module 401 for acquiring a value of each of a plurality of predictors of a target patient hospitalized for acute heart failure;

The searching module 402 is configured to search, from correspondence between a plurality of value intervals of each predictor and the readmission risk score, a readmission risk score corresponding to a numerical value of each predictor;

The adding module 403 is configured to add the readmission risk scores to obtain a total readmission risk score of the target patient;

The searching module 404 is further configured to search for a target readmission probability corresponding to the readmission risk total score of the target patient from a correspondence between the readmission risk total score and the readmission occurrence probability, so as to obtain a probability of readmission of the target patient.

According to the medical event probability determination device provided by the embodiment of the application, the readmission risk scores corresponding to the numerical values of each prediction index of the target patient are searched from the corresponding relations between the multiple value intervals of each prediction index of the acute heart failure and the readmission risk scores, and the readmission risk scores of the target patient are subjected to addition processing, so that the readmission risk total score of the target patient is obtained, namely, the target patient readmission risk total score is comprehensively determined according to the multiple prediction indexes, the target readmission probability corresponding to the target patient readmission risk total score is searched from the corresponding relations between the readmission risk total score and the readmission occurrence probability of the target patient, so that the readmission probability of the target patient is obtained, the accuracy of determining the readmission probability of the target patient is effectively improved, and in addition, the readmission probability of the target patient can be rapidly searched through the corresponding relations between the multiple value intervals of each prediction index and the readmission risk score, and the readmission occurrence probability of the target patient can be improved, and the clinical evaluation requirement of individual risk is met.

As an implementation manner of the present application, the foregoing apparatus may further include:

The acquisition module is also used for acquiring the value range of the first prediction index, the basic risk reference value of the first evaluation index and the constant value of the patient hospitalized by the acute heart failure; the first prediction index is any one of a plurality of prediction indexes, and the constant value is used for indicating the value corresponding to each 1 time-sharing increase of the risk score;

The determining module is used for dividing the value range into a plurality of value intervals and determining index reference values of each value interval;

The calculation module is used for calculating the distance between the index reference value and the basic risk reference value of each value interval;

The calculation module is also used for calculating the ratio between the distance and the constant value to obtain the readmission risk score corresponding to each value interval of the first prediction index;

the determining module is further configured to obtain a correspondence between the multiple value intervals of the first prediction index and the readmission risk score based on the value intervals of the first prediction index and the readmission risk scores corresponding to the value intervals of the first prediction index.

As an implementation of the present application, the search module 402 may include:

The determining submodule is used for determining a target value range corresponding to the numerical value of each prediction index from the value ranges of the multiple value intervals of each prediction index;

And the searching sub-module is used for searching the risk score corresponding to the target value range from the corresponding relation between the multiple value intervals of each prediction index and the readmission risk score to obtain the risk score corresponding to the numerical value of the prediction index.

The acquisition module is also used for acquiring a total readmission risk score of the first patient hospitalized with acute heart failure, a numerical value of each prediction index of the p prediction indexes, a regression coefficient of each prediction index, an average numerical value of each prediction index of the p prediction indexes of the patient hospitalized with acute heart failure and an average readmission probability; the first patient is any patient hospitalized for acute heart failure;

the determining module is further used for determining the sum of the basic risk value and the individual risk value of the first patient based on the numerical value of each prediction index of the p prediction indexes of the first patient and the regression coefficient of each prediction index;

A determination module for determining an average risk of readmission of the patient hospitalized for acute heart failure based on the average value of each of the p predictors for the patient hospitalized for acute heart failure and the regression coefficient of each predictor;

The determining module is further used for determining the readmission probability of the first patient based on the average readmission probability, the sum of the basic risk value and the individual risk value of the first patient and the average risk of crowd readmission;

The determining module is further configured to determine a correspondence between the total risk score of the first patient and a readmission probability of the first patient based on the total risk score of the first patient and the readmission probability.

the acquisition module is also used for acquiring a baseline risk function of a second patient hospitalized for acute heart failure at a target moment, the interval time from discharge to readmission and the numerical value of each prediction index of the p prediction indexes; the second patient is any patient hospitalized for acute heart failure;

A determination module further configured to determine a risk function value for the second patient at the target time based on the interval duration from discharge to readmission;

the determining module is further used for determining a regression coefficient of each prediction index based on the risk function value of each second patient at the target moment, the interval time from discharge to readmission and the numerical value of each prediction index of the p prediction indexes.

The determining module is also used for determining a risk index of readmission when the target patient is in the discharge preset time based on the probability of readmission of the target patient;

the determining module is further used for determining the readmission risk level of the target patient based on the risk index of readmission when the target patient is in the preset time after discharge.

Fig. 5 shows a schematic hardware structure of medical event probability determination according to an embodiment of the present application.

The medical event probability determination device may comprise a processor 501 and a memory 502 storing computer program instructions.

In particular, the processor 501 may include a Central Processing Unit (CPU), or an Application SPECIFIC INTEGRATED Circuit (ASIC), or may be configured as one or more integrated circuits that implement embodiments of the present application.

Memory 502 may include mass storage for data or instructions. By way of example, and not limitation, memory 502 may comprise a hard disk drive (HARD DISK DRIVE, HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or universal serial bus (Universal Serial Bus, USB) drive, or a combination of two or more of the foregoing. Memory 502 may include removable or non-removable (or fixed) media, where appropriate. Memory 502 may be internal or external to the integrated gateway disaster recovery device, where appropriate. In a particular embodiment, the memory 502 is a non-volatile solid state memory.

The memory may include Read Only Memory (ROM), random Access Memory (RAM), magnetic disk storage media devices, optical storage media devices, flash memory devices, electrical, optical, or other physical/tangible memory storage devices. Thus, in general, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software comprising computer-executable instructions and when the software is executed (e.g., by one or more processors) it is operable to perform the operations described with reference to methods in accordance with aspects of the present disclosure.

The processor 501 implements any of the medical event probability determination methods of the above embodiments by reading and executing computer program instructions stored in the memory 502.

In one example, the medical event probability determination device may further include a communication interface 503 and a bus 510. As shown in fig. 5, the processor 501, the memory 502, and the communication interface 503 are connected to each other by a bus 510 and perform communication with each other.

The communication interface 503 is mainly used to implement communication between each module, apparatus, unit and/or device in the embodiments of the present application.

The bus 510 includes hardware, software, or both that couple components of the medical event probability determination device to each other. By way of example, and not limitation, the buses may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an infiniband interconnect, a Low Pin Count (LPC) bus, a memory bus, a micro channel architecture (MCa) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a video electronics standards association local (VLB) bus, or other suitable bus, or a combination of two or more of the above. Bus 510 may include one or more buses, where appropriate. Although embodiments of the application have been described and illustrated with respect to a particular bus, the application contemplates any suitable bus or interconnect.

The medical event probability determining device can execute the medical event probability determining method in the embodiment of the application based on the currently intercepted junk short messages and the short messages reported by the user, thereby realizing the medical event probability determining method and the medical event probability determining device described in connection with fig. 1 to 3.

In addition, in combination with the medical event probability determination method in the above embodiment, the embodiment of the present application may be implemented by providing a computer storage medium. The computer storage medium has stored thereon computer program instructions; the computer program instructions, when executed by a processor, implement a medical event probability determination method of any of the above embodiments.

It should be understood that the application is not limited to the particular arrangements and instrumentality described above and shown in the drawings. For the sake of brevity, a detailed description of known methods is omitted here. In the above embodiments, several specific steps are described and shown as examples. The method processes of the present application are not limited to the specific steps described and shown, but various changes, modifications and additions, or the order between steps may be made by those skilled in the art after appreciating the spirit of the present application.

The functional blocks shown in the above-described structural block diagrams may be implemented in hardware, software, firmware, or a combination thereof. When implemented in hardware, it may be, for example, an electronic circuit, an Application Specific Integrated Circuit (ASIC), suitable firmware, a plug-in, a function card, or the like. When implemented in software, the elements of the application are the programs or code segments used to perform the required tasks. The program or code segments may be stored in a machine readable medium or transmitted over transmission media or communication links by a data signal carried in a carrier wave. A "machine-readable medium" may include any medium that can store or transfer information. Examples of machine-readable media include electronic circuitry, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio Frequency (RF) links, and the like. The code segments may be downloaded via computer networks such as the internet, intranets, etc.

It should also be noted that the exemplary embodiments mentioned in this disclosure describe some methods or systems based on a series of steps or devices. The present application is not limited to the order of the above-described steps, that is, the steps may be performed in the order mentioned in the embodiments, or may be performed in a different order from the order in the embodiments, or several steps may be performed simultaneously.

Aspects of the present disclosure are described above with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the disclosure. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions/acts specified in the flowchart and/or block diagram block or blocks. Such a processor may be, but is not limited to being, a general purpose processor, a special purpose processor, an application specific processor, or a field programmable logic circuit. It will also be understood that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware which performs the specified functions or acts, or combinations of special purpose hardware and computer instructions.

In the foregoing, only the specific embodiments of the present application are described, and it will be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the systems, modules and units described above may refer to the corresponding processes in the foregoing method embodiments, which are not repeated herein. It should be understood that the scope of the present application is not limited thereto, and any equivalent modifications or substitutions can be easily made by those skilled in the art within the technical scope of the present application, and they should be included in the scope of the present application.

Claims

1. A method of determining a probability of a medical event, comprising:

Searching a readmission risk score corresponding to the numerical value of each prediction index from the corresponding relation between a plurality of value intervals of each prediction index and the readmission risk score;

adding the readmission risk scores to obtain a readmission risk total score of the target patient;

Searching the target readmission probability corresponding to the target patient readmission risk total score from the corresponding relation between the readmission risk total score and the readmission occurrence probability, and obtaining the target patient readmission probability.

2. The medical event probability determination method according to claim 1, wherein before searching for a readmission risk score corresponding to the numerical value of each of the predictors from the correspondence between the plurality of value intervals of each predictor and the readmission risk score, the method further comprises:

Acquiring a value range of a first prediction index, a basic risk reference value of a first evaluation index and a constant value of a patient hospitalized for acute heart failure; the first prediction index is any one of a plurality of prediction indexes, and the constant value is used for indicating a value corresponding to each 1 time-sharing increase of the risk score;

And obtaining the corresponding relation between the multiple value intervals of the first prediction index and the readmission risk score based on the value intervals of the first prediction index and the corresponding readmission risk scores thereof.

3. The method according to claim 1, wherein searching for a readmission risk score corresponding to the value of each predictor from the correspondence between the plurality of value intervals of each predictor and the readmission risk score, comprises:

4. A medical event probability determination method according to any one of claims 1-3, wherein, from the correspondence between the total readmission risk score and the probability of readmission occurrence, the target readmission probability corresponding to the total readmission risk score of the target patient is searched for, and before obtaining the probability of readmission of the target patient, the method further comprises:

acquiring a total readmission risk score of a first patient hospitalized with acute heart failure, a numerical value of each of the P predictors, a regression coefficient of each predictor, an average numerical value of each predictor of the P predictors of a population hospitalized with acute heart failure, and an average readmission probability of the first patient hospitalized with acute heart failure; the first patient is any patient hospitalized for acute heart failure; p is an integer greater than 1;

determining the sum of a basic risk value and an individual risk value of a first patient based on the numerical value of each prediction index of P prediction indexes of the first patient and the regression coefficient of each prediction index;

determining an average risk of readmission of patients hospitalized for acute heart failure based on the average value of each of the P predictors for patients hospitalized for acute heart failure and the regression coefficient of each predictor;

Determining a readmission probability for the first patient based on the average readmission probability, a sum of the base risk value and the individual risk value for the first patient, and the average risk of crowd readmission;

5. The method of claim 4, wherein prior to the obtaining of the total readmission risk score for the first patient hospitalized for acute heart failure, the numerical value for each predictor of the P predictors, the regression coefficient for each predictor, the average numerical value for each predictor of the P predictors for the patient hospitalized for acute heart failure, and the average probability of readmission, the method further comprises:

Determining a risk function value of the second patient at a target time based on the interval duration from discharge to readmission;

And determining a regression coefficient of each prediction index based on the risk function value of each second patient at the target moment, the interval duration from discharge to readmission and the numerical value of each prediction index of the P prediction indexes.

6. A medical event probability determination method according to any one of claims 1-3, wherein the method further comprises:

Determining a risk index of readmission of the target patient in a preset time period after discharge of the target patient based on the probability of readmission of the target patient;

7. A medical event probability determination apparatus, the apparatus comprising:

the adding module is used for adding the readmission risk scores to obtain a readmission risk total score of the target patient;

The searching module is further used for searching the target readmission probability corresponding to the target patient readmission risk total score from the corresponding relation between the readmission risk total score and the readmission occurrence probability, and obtaining the target patient readmission probability.

8. A medical event probability determination device, the device comprising: a processor and a memory storing computer program instructions;

The processor, when executing the computer program instructions, implements the medical event probability determination method as defined in any one of claims 1-6.

9. A computer readable storage medium, characterized in that the computer readable storage medium has stored thereon computer program instructions, which when executed by a processor, implement the medical event probability determination method according to any of claims 1-6.

10. A computer program product, characterized in that instructions in the computer program product, when executed by a processor of an electronic device, cause the electronic device to perform the medical event probability determination method according to any of claims 1-6.