An information management system for asthma patients

Abstract

This application relates to the field of asthma analysis technology and provides an asthma patient information management system. This system includes an acquisition module for acquiring the values ​​of all influencing factors for multiple asthma patients of different types; a construction module for constructing a linear regression model between all influencing factors and hospital stay duration for each type of asthma patient, and a readmission risk model for each type of asthma patient; an analysis module for analyzing the predicted status of asthma patients based on the linear regression model and the readmission risk model, and obtaining the total hospital stay and average number of hospitalizations based on the predicted status; and a treatment recommendation module for outputting treatment recommendations for target asthma patients based on the total hospital stay and average number of hospitalizations for all types. This system can improve the rationality of treatment recommendations for asthma patients.

Description

Technical Field

[0001] This application relates to the field of asthma analysis technology, and in particular to an asthma patient information management system. Background Technology

[0002] Currently, there is insufficient control over readmissions and the use of medical resources in asthma management. On the one hand, there is a lack of precise analytical tools, making it difficult to accurately identify key factors affecting readmissions and the use of medical resources for asthma patients, thus hindering the development of targeted interventions. On the other hand, when simulating patient treatment processes and resource consumption, existing technologies cannot fully consider various complex factors, resulting in significant discrepancies between predicted and actual outcomes, and failing to provide a valid basis for treatment plan development.

[0003] Existing solutions: Some studies use simple statistical analyses to understand readmission rates in asthma patients, such as calculating the average readmission rate, but they cannot delve into the complex factors behind them. Regarding healthcare resource utilization, only simple linear models are established to analyze the relationship between a single factor (such as disease severity) and resource consumption, ignoring numerous other influencing factors.

[0004] The limitations of existing methods include: simple statistical analysis and linear models cannot comprehensively consider various factors such as individual patient differences, disease characteristics, treatment methods, medical policies, and lifestyle habits. These factors are intertwined and jointly influence the readmission rate of asthma patients. The one-sidedness of existing methods leads to inaccurate analysis results, which cannot provide reliable reference for medical decisions and result in unreasonable treatment recommendations for asthma patients. Summary of the Invention

[0005] This application provides an information management system for asthma patients, which can solve the problem of unreasonable treatment recommendations for asthma patients.

[0006] This application provides an asthma patient information management system, which includes:

[0007] The acquisition module is used to obtain the values ​​of all influencing factors for multiple asthma patients of different types; the influencing factors are the factors that affect the readmission of asthma patients, and each type corresponds to at least one asthma patient;

[0008] The module is used to build a linear regression model between all influencing factors and length of hospital stay for each type of asthma, based on the values ​​of all influencing factors for all asthma patients in that type, as well as a readmission risk model for asthma patients in that type.

[0009] The analysis module is used to analyze the predicted status of each asthma patient based on the linear regression model and readmission risk model corresponding to each asthma patient, and to obtain the total length of hospital stay and average number of hospitalizations for each type based on all predicted statuses.

[0010] The treatment recommendation module is used to output treatment recommendations for target asthma patients based on the total length of hospital stay and average number of hospitalizations for all types of patients.

[0011] Alternatively, the linear regression model is:

[0012]

[0013] For patients with the i-th type of asthma, Y i Indicates the length of hospital stay. L represents the intercept, and L represents the set of values ​​for all influencing factors. X represents the influence coefficient of factor l on the length of hospital stay. l Let i represent the value of influencing factor l, where i = 1, 2, ..., I, and I represents the number of patient types.

[0014] Optional readmission risk model:

[0015]

[0016] For patients with the i-th type of asthma, h(t|X) i Indicates the risk value for readmission. Represents the benchmark risk function. Let represent the influence coefficient of factor l on the readmission intensity, and exp represent the exponential function.

[0017] Optionally, based on the linear regression model and readmission risk model corresponding to each asthma patient, the predicted status of each asthma patient is analyzed, including:

[0018] The following steps will be performed separately for each asthma patient:

[0019] The values ​​of all influencing factors and the type of asthma patient are used as the current state;

[0020] A state transition function is constructed based on the current state, and the next state of the asthma patient and the time interval of state transition are calculated based on the state transition function and the readmission risk model. The patient state described by the next state is used as the predicted patient state of the asthma patient. The patient state is death or readmission. The state transition function is used to calculate the probability of the asthma patient transitioning to the next state and the probability of the time interval of state transition based on the values ​​of all influencing factors of the asthma patient and the type of asthma patient.

[0021] Determine whether the time interval is greater than the remaining observation time, or predict whether the patient's condition is death;

[0022] If the time interval is greater than the remaining observation time, or the patient's condition is predicted to be death, the number of times the patient's condition is predicted to be readmitted is counted, and the length of admission for each readmission is calculated according to the corresponding linear regression model. The number of readmissions and the total length of admission are used as the predicted condition for the patient's condition.

[0023] If the time interval is less than or equal to the remaining observation time and the predicted patient status is not death, then the next status is taken as the current status, the difference between the remaining observation time and the time interval is taken as the remaining observation time, and the process returns to the steps of constructing a state transition function based on the current status and calculating the predicted patient status of the asthma patient based on the state transition function.

[0024] Optionally, the state transition function is:

[0025]

[0026] Where p(s'|s,t) represents the probability that an asthma patient transitions from the current state s to state s' over a time interval t, and ω Death (s) indicates whether the asthma patient in the current state s has died; a value of 0 indicates not dead, and a value of 1 indicates dead. ω Death (s') indicates whether the asthma patient in state s' has died, L represents the set of values ​​for all influencing factors, ω l (s) represents the value of factor l in the current state s, ω l (s') represents the value of factor l in state s', ω Type (s) represents the type of asthma patient in the current state s, ω Type (s') represents the type of asthma patient in state s', ω Admission (s) represents the number of readmissions of asthma patients in the current state s, ω Admission (s') represents the number of readmissions of asthma patients in state s', and d and μ(s) represent the distribution parameters of the current state s, where d is the distribution parameter for whether the asthma patient has died, and μ(s) is the distribution parameter for the number of readmissions of asthma patients. The probability of readmission for asthma patients. Let F(s'|s,t) represent the probability of an asthma patient dying, and let F(s'|s,t) represent the probability of an asthma patient moving from s to s' without dying.

[0027]

[0028] in, Represents type ω Type (s) baseline readmission intensity for asthma patients, Represents ω Type (s) type patients' readmission and metastasis intensity parameters in relation to factor l Represents ω Type The rate of change parameter of the influencing factor l for patients of type (s) is f(s'|s,l,t), which represents the probability that the current state s and the state s' are consistent with feature l, and e represents the natural logarithm.

[0029] Optionally, the next state of the asthma patient and the time interval of the state transition are calculated based on the state transition function and the readmission risk model, including:

[0030] Multiple states and multiple time intervals are initialized based on the readmission risk model;

[0031] For each state, the transition from the current state to that state is calculated based on the state transition function, with a time interval of t. i The probability of t; i =1,2,...,T, where T represents the number of initialization time intervals;

[0032] The next state of an asthma patient is determined from all states and time intervals based on all probabilities, as well as the time interval for state transitions.

[0033] Optionally, the total length of stay and average number of hospitalizations for each type can be obtained based on all predicted states, including:

[0034] For each type, perform the following steps:

[0035] The total length of hospital stay for each type of asthma patient is obtained by summing the total length of hospital stay in the predicted state of all patients of that type.

[0036] The average number of readmissions for each type of asthma patient in the predicted state is calculated to obtain the average number of hospitalizations for that type.

[0037] The above-mentioned solution in this application has the following beneficial effects:

[0038] In some embodiments of this application, by obtaining the values ​​of all influencing factors for multiple asthma patients of different types, and then, for each type, constructing a linear regression model between all influencing factors and hospital stay duration, as well as a readmission risk model for asthma patients of that type, based on the values ​​of all influencing factors for all asthma patients of that type, and then analyzing the predicted status of each asthma patient based on the corresponding linear regression model and readmission risk model, and obtaining the total hospital stay and average number of hospitalizations for each type based on all predicted statuses, and finally outputting treatment recommendations for the target asthma patient based on the total hospital stay and average number of hospitalizations for all types. This comprehensive consideration of multiple influencing factors improves the rationality and practicality of the analysis of asthma patients, and the prediction of readmissions and hospital stay durations provides intuitive and reliable data on the number of hospitalizations and hospital stay durations, providing accurate data for the hospitalization and treatment management of asthma patients, thereby improving the rationality of the treatment recommendations for asthma patients.

[0039] Other beneficial effects of this application will be described in detail in the following detailed description section. Attached Figure Description

[0040] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 This is a schematic diagram of the structure of an asthma patient information management system provided in an embodiment of this application;

[0042] Figure 2 A flowchart illustrating the usage process of an asthma patient information management system provided in an embodiment of this application;

[0043] Figure 3 A simulation flowchart of a single patient provided in one embodiment of this application;

[0044] Figure 4 A simulation flowchart of a population provided in one embodiment of this application. Detailed Implementation

[0045] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0046] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0047] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0048] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0049] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0050] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0051] To address the issue of inadequate existing treatment and management plans for asthma patients, this application provides an asthma patient information management system. This system acquires the values ​​of all influencing factors for multiple asthma patients of different types. Then, for each type, based on the values ​​of all influencing factors for all asthma patients within that type, it constructs a linear regression model between all influencing factors and hospital stay duration, as well as a readmission risk model for asthma patients within that type. Based on the linear regression model and readmission risk model corresponding to each asthma patient, it analyzes and obtains the predicted status of each asthma patient. Based on all predicted statuses, it calculates the total hospital stay duration and average number of hospitalizations for each type. Finally, based on the total hospital stay duration and average number of hospitalizations for all types, it outputs treatment recommendations for target asthma patients. By comprehensively considering multiple influencing factors, it improves the rationality and practicality of asthma patient analysis. The system predicts the readmission and hospital stay status of asthma patients, providing intuitive and reliable data on the number of hospitalizations and hospital stay duration, offering accurate data for the hospitalization and treatment management of asthma patients, thereby improving the rationality of treatment recommendations for asthma patients.

[0052] like Figure 1 As shown, the asthma patient information management system 100 provided in this application includes:

[0053] The acquisition module 101 is used to acquire the values ​​of all influencing factors for multiple asthma patients of different types; the influencing factors are the factors that affect the readmission of asthma patients, and each type corresponds to at least one asthma patient;

[0054] Module 102 is used to construct, for each type, a linear regression model between all influencing factors and length of hospital stay under each type, and a readmission risk model for asthma patients under each type, based on the values ​​of all influencing factors for all asthma patients under each type.

[0055] Analysis module 103 is used to analyze the predicted status of each asthma patient based on the linear regression model and readmission risk model corresponding to each asthma patient, and to obtain the total length of hospital stay and average number of hospitalizations for each type based on all predicted statuses.

[0056] The treatment recommendation module 104 is used to output treatment recommendation information for target asthma patients based on the total length of hospital stay and average number of hospitalizations for all types of patients.

[0057] like Figure 2 As shown, the asthma patient information management system provided in this application includes the following steps during use:

[0058] Step 21: Obtain the values ​​of all influencing factors for multiple asthma patients of different types.

[0059] The above-mentioned influencing factors affect the readmission rate of asthma patients (such as the number of previous hospitalizations, survival status, status of intermediate-acting corticosteroid use, status of long-acting corticosteroid use, status of antibiotic use, status of β2-receptor agonist use, status of leukotriene receptor antagonist use, etc.), with each type corresponding to at least one asthma patient. Types are categorized based on the asthma patient's gender, medical insurance type, and primary diagnosis, such as mild asthma, moderate asthma, and severe asthma.

[0060] In some embodiments of this application, the values ​​of influencing factors for asthma patients can be obtained by accessing a hospital's medical system or similar means.

[0061] It should be noted that for influencing factors whose data is text-based, corresponding values ​​can be obtained by encoding them. For example, survival status is represented by code 1 for death and code 0 for survival, and antibiotic use status is represented by code 1 for taking the medication and code 0 for not taking the medication.

[0062] Step 22: For each type, based on the values ​​of all influencing factors for all asthma patients in that type, construct a linear regression model between all influencing factors and length of hospital stay for that type, as well as a readmission risk model for asthma patients in that type.

[0063] Specifically, the linear regression model is as follows:

[0064]

[0065] For patients with the i-th type of asthma, Y i Indicates the length of hospital stay. L represents the intercept, and L represents the set of values ​​for all influencing factors. X represents the influence coefficient of factor l on the length of hospital stay. l Let i represent the value of influencing factor l, where i = 1, 2, ..., I, and I represents the number of patient types.

[0066] The readmission risk model is as follows:

[0067]

[0068] For patients with the i-th type of asthma, h(t|X) i Indicates the risk value for readmission. Represents the benchmark risk function. Let represent the influence coefficient of factor l on the readmission intensity, and exp represent the exponential function.

[0069] It should be noted that the influence coefficients of the aforementioned influencing factors on hospital stay duration and readmission intensity can be obtained by collecting the values ​​of all influencing factors and hospital stay duration from multiple asthma patients as samples, and performing regression analysis on the values ​​of influencing factors and hospital stay duration. For the influence coefficient of influencing factors on hospital stay duration, hospital stay duration is used as the dependent variable and the values ​​of influencing factors for asthma patients are used as independent variables, and linear regression analysis is performed to obtain the influence coefficient of influencing factors on hospital stay duration. For the influence coefficient of influencing factors on readmission intensity, readmission risk value is used as the dependent variable and influencing factor values ​​for asthma patients are used as independent variables, and proportional hazard regression analysis is performed to obtain the influence coefficient of influencing factors on readmission intensity. The aforementioned baseline risk function is a readmission risk model when all independent variables (i.e., the values ​​of all influencing factors) are 0, considering only the relationship between time and readmission risk value, and is estimated using proportional hazard.

[0070] Step 23: Based on the linear regression model and readmission risk model corresponding to each asthma patient, analyze the predicted status of each asthma patient, and obtain the total length of hospital stay and average number of hospitalizations for each type based on all predicted statuses.

[0071] In some embodiments of this application, the steps of analyzing and obtaining the predicted status of each asthma patient based on the linear regression model and readmission risk model corresponding to each asthma patient include:

[0072] The following steps will be performed separately for each asthma patient:

[0073] The first step is to set the values ​​of all influencing factors and the type of asthma patient as the current state.

[0074] The second step is to construct a state transition function based on the current state, and then calculate the next state of the asthma patient and the time interval of the state transition based on the state transition function and the readmission risk model. The patient state described by the next state is used as the predicted patient state of the asthma patient.

[0075] The patient's status is either death or readmission. The state transition function is used to calculate the probability of an asthma patient transitioning to the next state and the probability of the time interval between state transitions, based on the values ​​of all influencing factors and the patient's type. The current state refers to the asthma patient's state at the current moment, and the next state is the state of the asthma patient at the moment following the current moment.

[0076] Specifically, the state transition function is:

[0077]

[0078] Where p(s'|s,t) represents the probability that an asthma patient transitions from the current state s to state s' over a time interval t, and ω Death (s) indicates whether the asthma patient in the current state s has died; a value of 0 indicates not dead, and a value of 1 indicates dead. ω Death (s') indicates whether the asthma patient in state s' has died, L represents the set of values ​​for all influencing factors, ω l (s) represents the value of factor l in the current state s, ω l (s') represents the value of factor l in state s', ω Type (s) represents the type of asthma patient in the current state s, ω Type (s') represents the type of asthma patient in state s', ω Admission (s) represents the number of readmissions of asthma patients in the current state s, ω Admission (s') represents the number of readmissions of asthma patients in state s', and d and μ(s) represent the distribution parameters of the current state s, where d is the distribution parameter for whether the asthma patient has died, and μ(s) is the distribution parameter for the number of readmissions of asthma patients. The probability of readmission for asthma patients. Let F(s'|s,t) represent the probability of an asthma patient dying, and let F(s'|s,t) represent the probability of an asthma patient moving from s to s' without dying.

[0079]

[0080] in, Represents type ω Type (s) baseline readmission intensity for asthma patients, Represents ω Type (s) type patients' readmission and metastasis intensity parameters in relation to factor l Represents ω Type The rate of change parameter of influencing factor l for patients with type (s) is the coefficient (intensity) of change of influencing factor l over time. f(s'|s,l,t) represents the probability that the current state s and state s' are consistent with feature l. e represents the natural logarithm and t represents time t, which is used to describe the time step of state transition in this step.

[0081] It should be noted that the above state transition function assumes that the state transition processes of each feature are independent, which is realistic and easy to calculate. In this algorithm, the readmission intensity μ(s) is independent of time and only depends on the patient's current state; μ(s) is a special case of h(t|X). correspond It is a constant that is only related to patient type and is independent of time. and Correspondingly; I(ω) Death (s) = 0) is used to characterize that the re-admission intensity is 0 if the patient has already died.

[0082] The steps described above for calculating the next state and the time interval of state transition for asthma patients based on the state transition function and readmission risk model are as follows:

[0083] First, multiple states and multiple time intervals are initialized based on the readmission risk model.

[0084] Specifically, the readmission risk model reflects the risk of asthma patients being readmitted, and to some extent, the probability of readmission. The model calculates the readmission risk value by substituting the patient's current state into the model, and then generates multiple states and time intervals based on this value. For example, the readmission risk value can be used to filter the randomly generated states; if the readmission risk value is 0.5, then there is a 50% probability of readmission, and half of the generated states should correspond to a readmission patient state. Multiple time intervals can be randomly generated. Each state includes the values ​​of all random influencing factors, as well as the patient's state (death or readmission).

[0085] Then, for each state, the transition from the current state to that state is calculated according to the state transition function, with a time interval of t. i The probability of t; i =1,2,...,T, where T represents the number of initialization time intervals.

[0086] Specifically, each state and time interval is substituted into the state transition function to calculate the corresponding probability. That is, each state corresponds to T probabilities.

[0087] Finally, the next state of the asthma patient, and the time interval for state transition, are determined from all states and time intervals based on all probabilities. For example, in the current state A, the probability of transitioning to the next state B1 with a time interval of C1 is 0.2, the probability of transitioning to the next state B2 with a time interval of C2 is 0.5, and the probability of transitioning to the next state B3 with a time interval of C3 is 0.3. That is, there is a 0.2 probability of transitioning to state B1 with a time interval of C1, a 0.5 probability of transitioning to state B2 with a time interval of C2, and a 0.3 probability of transitioning to state B3 with a time interval of C3.

[0088] The third step is to determine whether the time interval is greater than the remaining observation period, or to predict whether the patient's condition is death.

[0089] The remaining observation time mentioned above is the remaining observation time for asthma patients. The initial value of the observation time is a preset value, which is set according to the actual duration of asthma illness of the patients.

[0090] If the time interval is greater than the remaining observation time, or the patient's condition is predicted to be death, the number of times the patient's condition is predicted to be readmitted is counted, and the length of admission for each readmission is calculated according to the corresponding linear regression model. The number of readmissions and the total length of admission are used as the predicted condition for the patient's condition.

[0091] If the time interval is less than or equal to the remaining observation time and the predicted patient status is not death, then the next status is taken as the current status, the difference between the remaining observation time and the time interval is taken as the remaining observation time, and the process returns to the steps of constructing a state transition function based on the current status and calculating the predicted patient status of the asthma patient based on the state transition function.

[0092] Specifically, the values ​​of all influencing factors at each readmission are substituted into the linear regression model to obtain the readmission duration.

[0093] The fourth step is to obtain the total length of stay and average number of hospitalizations for each type based on all predicted states.

[0094] Specifically, for each type, the following steps are performed:

[0095] The total length of hospital stay for each type is obtained by summing the total length of hospital stay in the predicted state of all asthma patients in the corresponding type.

[0096] The average number of readmissions for each type of asthma patient in the predicted state is calculated to obtain the average number of hospitalizations for that type.

[0097] It should be noted that this step can also incorporate the medical resources consumed by asthma patients during each hospitalization for calculation. The medical resource consumption data of asthma patients is added to the current state and also introduced during the initialization state. Through the calculation process described above, the medical resource consumption status of each type of asthma patient can be obtained, such as the average medical resources consumed by this type of patient during each hospitalization and the total medical resource consumption of this type of patient.

[0098] The following example illustrates this step.

[0099] like Figure 3As shown, the simulation software is used to implement the processing of a single asthma patient in this step. After the patient enters the simulation system, the duration of the first hospitalization is recorded, and the next event (used to describe whether the patient is dead or readmitted, equivalent to the state mentioned above) and the time interval T are generated. Then, it is determined whether T is greater than the remaining observation period (i.e., the remaining observation duration mentioned above). If so, the total hospitalization duration and the number of hospitalizations within the observation period are calculated. Otherwise, the event type is determined. If it is death, the total hospitalization duration and the number of hospitalizations within the observation period are calculated. If it is readmitted, the remaining observation period is updated to the original remaining period - T, and the duration of this hospitalization is recorded. The process then returns to the step of generating the next event and the event interval T.

[0100] The simulation process for all asthma patients is as follows: Figure 4 As shown, after starting, the patient type and mortality intensity parameters are determined. Then, the patient set of that type is initialized, and it is determined whether there are any untreated patients. If so, the next patient is retrieved, and the process is executed. Figure 3 The simulation shown targets a single asthma patient, records the patient's total hospital stay and number of hospitalizations, and returns to the step of determining whether there are any untreated patients. If there are no untreated patients, the data of all patients are aggregated, the total hospital stay and average number of hospitalizations for each group are calculated, and the process ends.

[0101] Step 24: Based on the total length of hospital stay and average number of hospitalizations for all types of patients, output treatment recommendations for the target asthma patient.

[0102] The aforementioned target asthma patients are those who require treatment.

[0103] For example, to facilitate analysis, the total length of hospital stay and average number of hospitalizations for each type can be converted into values ​​commonly used in statistical analysis, such as mean and standard deviation. Based on the type of the target asthma patient, the corresponding values ​​are used to determine whether the target asthma patient needs readmission, or to provide treatment recommendations such as the length of readmission. For instance, if the target asthma patient has mild asthma, the average number of hospitalizations for this type is 4, and the target asthma patient has only been hospitalized once, then the probability of subsequent readmission is considered high, and treatment recommendations for readmission at any time are obtained.

[0104] It should be noted that medical personnel can also formulate reasonable medical resource allocation plans and personalized treatment plans for target patients based on the total length of hospital stay, average number of hospitalizations, and medical resource consumption for all types of patients, thereby improving the rationality and reliability of patient treatment management.

[0105] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0106] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0107] The above description is the preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principles described in this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. An information management system for asthma patients, characterized in that, include: The acquisition module is used to obtain the values ​​of all influencing factors for multiple asthma patients with different types of diseases. The influencing factors are those that affect the readmission of asthma patients, and each type corresponds to at least one asthma patient; The module is used to construct, for each of the aforementioned types, a linear regression model between all influencing factors and length of hospital stay under the aforementioned type, and a readmission risk model for asthma patients under the aforementioned type, based on the values ​​of all influencing factors for all asthma patients under the aforementioned type. The analysis module is used to analyze the predicted status of each asthma patient based on the linear regression model and readmission risk model corresponding to each asthma patient, and to obtain the total length of hospital stay and the average number of hospitalizations for each type based on all predicted statuses. The treatment recommendation module is used to output treatment recommendation information for target asthma patients based on the total length of hospital stay and average number of hospitalizations for all types of patients. Specifically, the analysis module is used to implement: The following steps will be performed separately for each asthma patient: The values ​​of all influencing factors of the asthma patient and the type of the asthma patient are used as the current state; A state transition function is constructed based on the current state, and the next state and the time interval of the state transition are calculated based on the state transition function and the readmission risk model. The patient state described by the next state is used as the predicted patient state of the asthma patient. The patient state is death or readmission. The state transition function is used to calculate the probability of the asthma patient transitioning to the next state and the probability of the time interval of the state transition based on the values ​​of all influencing factors of the asthma patient and the type of the asthma patient. Determine whether the time interval is greater than the remaining observation time, or whether the predicted patient status is death; If the time interval is greater than the remaining observation time, or the predicted patient status is death, then the number of times the predicted patient status is readmission is counted, and the admission time for each readmission is calculated according to the corresponding linear regression model. The number of readmissions and the total admission time are used as the predicted status of the predicted patient status. If the time interval is less than or equal to the remaining observation duration and the predicted patient state is not death, then the next state is taken as the current state, the difference between the remaining observation duration and the time interval is taken as the remaining observation duration, and the process returns to the step of constructing a state transition function based on the current state and calculating the predicted patient state of the asthma patient based on the state transition function.

2. The asthma patient information management system according to claim 1, characterized in that, The linear regression model is as follows: For patients with the i-th type of asthma, Y i Indicates the length of hospital stay. L represents the intercept, and L represents the set of values ​​for all influencing factors. X represents the influence coefficient of factor l on the length of hospital stay. l Let i represent the value of influencing factor l, where i = 1, 2, ..., I, and I represents the number of patient types.

3. The asthma patient information management system according to claim 2, characterized in that, The readmission risk model is as follows: For patients with the i-th type of asthma, h(t|X) i This indicates the risk value for readmission. Represents the benchmark risk function. Let represent the influence coefficient of factor l on the readmission intensity, and exp represent the exponential function.

4. The asthma patient information management system according to claim 1, characterized in that, The state transition function is: Where p(s'|s,t) represents the probability that an asthma patient transitions from the current state s to state s' over a time interval t, and ω Death (s) indicates whether the asthma patient in the current state s has died; a value of 0 indicates not dead, and a value of 1 indicates dead. ω Death (s') indicates whether the asthma patient in state s' has died, L represents the set of values ​​for all influencing factors, ω l (s) represents the value of factor l in the current state s, ω l (s') represents the value of factor l in state s', ω Type (s) represents the type of asthma patient in the current state s, ω Type (s') represents the type of asthma patient in state s', ω Admission (s) represents the number of readmissions of asthma patients in the current state s, ω Admission (s') represents the number of readmissions of asthma patients in state s', and d and μ(s) represent the distribution parameters of the current state s, where d is the distribution parameter for whether the asthma patient has died, and μ(s) is the distribution parameter for the number of readmissions of asthma patients. The probability of readmission for asthma patients. Let F(s'|s,t) represent the probability of an asthma patient dying, and let F(s'|s,t) represent the probability of an asthma patient moving from s to s' without dying. in, Represents type ω Type (s) baseline readmission intensity for asthma patients, Represents ω Type (s) type patients' readmission and metastasis intensity parameters in relation to factor l Represents ω Type The rate of change parameter of the influencing factor l for patients of type (s) is f(s'|s,l,t), which represents the probability that the current state s and the state s' are consistent with feature l, and e represents the natural logarithm.

5. The asthma patient information management system according to claim 4, characterized in that, The calculation of the next state of the asthma patient and the time interval of state transition based on the state transition function and the readmission risk model includes: Multiple states and multiple time intervals are initialized based on the readmission risk model; For each of the states, the transition from the current state to the stated state is calculated based on the state transition function, with a time interval of t. i The probability of t; i =1,2,...,T, where T represents the number of initialization time intervals; The next state of the asthmatic patient is determined from all states and time intervals based on all probabilities, as well as the time interval for state transition.

6. The asthma patient information management system according to claim 1, characterized in that, The process of obtaining the total length of stay and average number of hospitalizations for each type based on all predicted states includes: For each of the aforementioned types, the following steps are performed: The total length of hospital stay for the specified type is obtained by summing the total length of hospital stay in the predicted state of all asthma patients corresponding to the specified type. The average number of readmissions for all asthma patients corresponding to the aforementioned type is obtained by averaging the number of readmissions in the predicted state.

Images