Pre-hospital registration method for intelligent management planning

Abstract

The invention discloses a pre-hospital registration method for intelligent management planning, and relates to the field of hospitalization management, and the method comprises the following steps: 1, obtaining the access authority of a hospital electronic health recording system, and reading the basic personal information and past medical history of a target patient; 2, recording current personal symptoms and personal signs of the target patient in real time, and obtaining a treatment scheme made by a current doctor for the target patient; 3, constructing an analysis model through a deep learning algorithm, obtaining historical medical record data through a hospital electronic health recording system, and extracting tag features of personal information, medical history, symptoms, signs, treatment schemes, hospitalization time and treatment effects of a plurality of patients; through multi-dimensional analysis of basic information, medical history, symptoms and treatment schemes of a patient, the analysis model can generate a more accurate illness condition severity coefficient, and through combination of the illness condition severity coefficient and real-time information of a nursing bed, an appropriate bed and a nursing level are automatically matched.

Description

Technical Field

[0001] The present invention relates to the technical field of hospitalization management, and in particular to a pre-hospital registration method for intelligent management planning. Background Art

[0002] With the aging of the population and the increase in the prevalence of chronic diseases, the demand for medical services has risen sharply. At the same time, patients' conditions are complex and require personalized medical plans, which poses a major challenge to hospitals in effectively managing patients and medical resources. The classification and treatment of emergency cases and non-emergency cases before hospitalization is particularly important to ensure that critically ill patients can receive timely treatment; In recent years, many hospitals have implemented electronic health record systems, which enable digital management of patient information and promote information sharing and exchange. Many countries and regions are carrying out medical reforms and encouraging the use of intelligent technology, data analysis and information technology to improve medical quality and efficiency. The rapid progress of modern medicine has led to the rapid update of medical knowledge. When doctors face complex cases, they often need more efficient and scientific decision-making support. Existing inpatient registration systems mostly use universal assessment standards and fail to provide personalized management based on individual differences and disease characteristics of patients, which can easily lead to misdiagnosis or delayed treatment. Hospitals often lack a basis for allocating beds and nursing resources, resulting in waste of resources or insufficient resources in critical situations. Doctors lack effective decision-making support tools when faced with complex cases and are easily influenced by subjective factors and ignore certain important information. Summary of the invention

[0003] 1. Technical issues to be resolved In view of the above-mentioned shortcomings of the prior art, the present invention provides a pre-hospital registration method with intelligent management planning, which can effectively solve the problems of the prior art.

[0004] (II) Technical solution To achieve the above objectives, the present invention is implemented through the following technical solutions: The present invention discloses a pre-hospital registration method for intelligent management planning, comprising the following steps: Step 1: Obtain access to the hospital's electronic health record system to read the target patient's basic personal information and medical history; Step 2: Record the current personal symptoms and signs of the target patient in real time, and obtain the treatment plan currently formulated by the doctor for the target patient; Step 3: Use deep learning algorithms to build an analysis model, obtain historical medical record data through the hospital's electronic health record system, and extract label features of several patients' personal information, medical history, symptoms, signs, treatment plans, hospitalization time, and treatment effects, build a training sample set, and input it into the analysis model for training; Step 4: Obtain the trained analysis model, extract the label features of the current target patient's personal information, medical history, symptoms, and treatment plan, form a label feature set, input the target patient's label feature set into the analysis model, and output the target patient's disease severity coefficient; Step 5: Construct a strategy library of patient condition severity coefficients and hospitalization plans corresponding to different label feature combinations, and match the hospitalization plan required by the current target patient in the strategy library based on the target patient condition severity coefficient currently output; Step 6: Enter the attributes of all beds of different nursing levels in the current hospital; Step 7: According to the patient's condition severity coefficient and the matched hospitalization plan, the nursing level beds required by the current patient are sorted and allocated, with priority given to patients with serious conditions, and a bed allocation list is generated and submitted to the management end.

[0005] Furthermore, the analysis model in step 3 represents the severity coefficient through a multiple linear regression strategy, and the calculation formula of the analysis model is: ; In the formula, Represents the severity coefficient, which indicates the severity of the target patient's condition. The larger the value, the more serious the condition. represents the intercept term, which indicates the baseline severity coefficient when all eigenvalues ​​are zero, represents the number of label features, Represents the coefficient of linear regression, indicating the characteristics The degree of influence on the severity coefficient S, represents the value of the i-th feature, Represents the number of features used to calculate nonlinear effects, Represents the nonlinear regression coefficient, indicating the characteristics The degree of influence of self-influence on the severity of the disease, represents the square of the jth feature, indicating the nonlinear effect of this feature on the severity of the disease. Represents the number of features and is used to calculate the interaction effects between features. Represents the interaction coefficient, indicating the characteristics With features The effect of the interaction term on the severity coefficient, Represents the product of the value of feature k and the value of feature l, indicating the interaction between feature k and feature l. represents the error term, which represents the partial or random error that the model cannot explain. It is assumed to follow a normal distribution with mean zero and is used for the summation of linear terms.

[0006] Furthermore, the analysis model calculation formula Captures the direct impact of each feature on the severity coefficient, indicating that the contribution of each feature to the severity coefficient is linear, through the square term of the feature Reflecting the increase in certain characteristic degrees, the impact on the disease is enhanced, through The interaction between features was analyzed to show the effect of the combination of two specific features on the severity coefficient.

[0007] Furthermore, the working logic of the analysis model calculation formula is: Collect relevant information and label characteristics of target patients; Select several important features based on the availability and relevance of information; A feature matrix is ​​constructed based on the selected features, and the feature values ​​of each patient form a vector; Use historical patient data to fit the model, and determine the degree of influence of the characteristics on the severity coefficient of the disease by fitting the model; Using the fitted parameters and the new patient characteristic values, calculate the severity coefficient; Interpret the calculated severity coefficient and judge the severity of the patient's condition. The higher the severity coefficient, the more serious the condition. Make clinical decisions based on the results of the severity coefficient. Continuously monitor and evaluate the performance of analytical models, validate them with new patient data, and update analytical models.

[0008] Furthermore, the process of matching the hospitalization plan required by the current target patient in the strategy library in step 5 is: A hospitalization strategy library containing detailed information of a plurality of hospitalization plans is predefined, and each hospitalization plan corresponds to a different severity coefficient interval and patient characteristics; Using the target patient's disease severity coefficient as a key indicator, search the hospitalization plans in the strategy library, and use conditional judgment or query statements to determine which hospitalization plans have the disease severity coefficient within the condition range; Output the hospitalization plan whose severity coefficient meets the condition range.

[0009] Furthermore, in step 3, the training sample set is divided into 70% training set and 30% validation set, and the ratio of the training set to the validation set is adjusted by receiving external instructions.

[0010] Furthermore, in step 5, if the target patient's condition severity coefficient currently output does not match the hospitalization plan required by the current target patient in the current policy library, the indexing is terminated, and a supplementary report of the associated data of the condition severity coefficient is generated and submitted to the management end.

[0011] Furthermore, the bed attributes in step 6 include: bed status, number of beds, nursing level and regional location.

[0012] Furthermore, the working logic of the sorting and allocation in step 7 is: traverse the sorted patient list, check the available beds according to the required nursing level of the current patient, and allocate corresponding beds to eligible patients. If the corresponding bed is full, skip the bed and continue to check the next bed. Once a bed is allocated to the patient, update the mark of the bed as occupied and update the available bed information. If a bed is not allocated to the patient, send a prompt message to the management end.

[0013] Furthermore, the bed allocation list in step 7 includes: patient identity information, allocated bed information, nursing level, hospitalization time information and disease severity coefficient.

[0014] (III) Beneficial effects Compared with the known prior art, the technical solution provided by the present invention has the following beneficial effects: Using deep learning algorithms and analytical models, the patient's condition is accurately assessed. Through multi-dimensional analysis of the patient's basic information, medical history, symptoms and treatment plan, the analytical model can generate a more accurate disease severity coefficient. Combined with the disease severity coefficient and real-time information of nursing beds, it automatically matches the appropriate beds and nursing levels. By giving priority to allocating relevant resources to patients with more serious conditions, the hospital's bed resources are maximized, which helps to improve the overall quality and efficiency of nursing care, improve the accuracy of the hospital's initial triage, and provide personalized medical plans based on the patient's specific condition, thereby improving treatment outcomes.

[0015] Through automated data collection and processing, medical staff can reduce the time spent on patient registration and information entry. Real-time recording of patients' symptoms and signs allows information to be captured and analyzed quickly, reducing patient waiting time, improving the response speed of medical services, and enabling doctors to make treatment decisions more quickly.

[0016] Based on real-time data and analysis results, medical resources and processes can be flexibly adjusted, and analytical models are continuously updated during training and verification to adapt to emerging patient characteristics and trends, enhancing the hospital's ability to respond to emergencies and improving the resilience of the medical system. It also ensures that the hospital can continuously optimize services based on the latest medical knowledge and technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 It is a schematic diagram of the process of the present invention; Figure 2 It is a flowchart of the working logic of the calculation formula of the analysis model in the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0020] The present invention will be further described below in conjunction with the embodiments.

[0021] Embodiment 1: A pre-hospital registration method for intelligent management planning in this embodiment, such as Figure 1 As shown, the following steps are included: Step 1: Obtain access to the hospital's electronic health record system and read the target patient's basic personal information and medical history. By accurately reading the target patient's basic personal information and medical history, the patient's health status can be quickly identified, providing basic information for subsequent processing. Compared with the traditional method of manually recording patient information, automated data acquisition reduces human errors and improves the accuracy and timeliness of information. Step 2: Record the current personal symptoms and signs of the target patient in real time, and obtain the treatment plan currently formulated by the doctor for the target patient; Record the current symptoms and signs in real time, so that the doctor can formulate an appropriate treatment plan based on a full understanding of the patient's condition. Compared with the traditional method that only relies on the doctor's subjective judgment, this step introduces real-time data recording, reduces forgetfulness and bias, and thus improves the accuracy of diagnosis; Step 3: Use deep learning algorithms to build an analysis model, obtain historical medical record data through the hospital's electronic health record system, and extract label features of several patients' personal information, medical history, symptoms, signs, treatment plans, hospitalization time, and treatment effects, build a training sample set, and input it into the analysis model for training; the training sample set is divided into 70% training set and 30% validation set, and the ratio of training set and validation set is adjusted by receiving external instructions; by analyzing a large amount of historical medical record data, the model can accurately extract important features and train to better evaluate and predict the patient's condition, which is significantly better than the previous evaluation system that relies on simple statistical methods. Deep learning can handle more complex features and find potential patterns, providing more accurate condition assessment; Step 4: Obtain the trained analysis model, extract the label features of the current target patient's personal information, medical history, symptoms, and treatment plan, form a label feature set, input the target patient's label feature set into the analysis model, and output the target patient's disease severity coefficient; the analysis model can automatically generate the patient's disease severity coefficient, objectively reflecting the patient's current health status. Compared with the traditional scoring system, this method has a high degree of automation, reduces the subjectivity of manual scoring, and makes the evaluation results more objective and accurate; Step 5: Construct a strategy library of patient severity coefficients and hospitalization plans corresponding to different label feature combinations. Based on the target patient severity coefficient currently output, match the hospitalization plan required by the current target patient in the strategy library. The process of matching the hospitalization plan required by the current target patient in the strategy library is as follows: A hospitalization strategy library containing detailed information of a plurality of hospitalization plans is predefined, and each hospitalization plan corresponds to a different severity coefficient interval and patient characteristics; Using the target patient's disease severity coefficient as a key indicator, search the hospitalization plans in the strategy library, and use conditional judgment or query statements to determine which hospitalization plans have the disease severity coefficient within the condition range; Output the hospitalization plan whose severity coefficient meets the condition range; If the target patient's condition severity coefficient currently output does not match the hospitalization plan required by the current target patient in the current strategy library, the index is terminated, and a supplementary report of the associated data of the condition severity coefficient is generated and submitted to the management end; the appropriate hospitalization plan is automatically matched according to different condition severity coefficients, which improves the efficiency of admission decision-making. This step enables the hospital to quickly respond to patient needs and actively formulate personalized hospitalization plans. Compared with the inefficiency and errors of manual selection of plans, it can provide services to patients more quickly and scientifically; Step 6: Enter the attributes of all beds of different nursing levels in the current hospital; bed attributes include: bed status, number of beds, nursing level and regional location; comprehensively record the attributes of nursing level beds to facilitate subsequent management and bed allocation. Compared with the manual management method of bed information, the electronic and structured management method of bed information greatly improves bed utilization and management efficiency and reduces the risk of information loss; Step 7: According to the patient's severity coefficient and the matched hospitalization plan, the nursing level beds required by the current patient are sorted and allocated, and bed allocation is performed first for patients with severe conditions, and a bed allocation list is generated and submitted to the management end. The working logic of sorting and allocation is as follows: traverse the sorted patient list, check the available beds according to the required nursing level of the current patient, and allocate the corresponding beds to eligible patients. If the corresponding bed is full, skip the bed and continue to check the next bed. Once a bed is allocated to the patient, update the mark of the bed as occupied and update the available bed information. If a bed is not allocated to the patient, send a prompt message to the management end; the bed allocation list includes: patient identity information, allocated bed information, nursing level, hospitalization time information and severity coefficient; according to the patient's severity coefficient, intelligently sort and allocate beds to ensure that patients with severe conditions receive priority care. Compared with traditional manual allocation, the automated allocation logic in this step avoids resource waste and unreasonable allocation, and improves bed utilization efficiency and patient admission experience.

[0022] In the specific implementation of this embodiment, a deep learning algorithm and an analysis model are used to accurately assess the patient's condition. Through a multi-dimensional analysis of the patient's basic information, medical history, symptoms, and treatment plan, the model can generate a more accurate disease severity coefficient. Combined with the disease severity coefficient and real-time information of nursing beds, the model automatically matches the appropriate beds and nursing levels. By giving priority to the allocation of relevant resources to patients with more serious conditions, the hospital's bed resources are maximized, which helps to improve the overall quality and efficiency of nursing care, improve the accuracy of the hospital's initial triage, and provide personalized medical plans based on the patient's specific condition, thereby improving the treatment effect. Through automated data collection and processing, the time spent by medical staff on patient registration and information entry is reduced. Real-time recording of patients' symptoms and signs allows information to be quickly captured and analyzed, reducing patient waiting time, improving the response speed of medical services, and enabling doctors to make treatment decisions more quickly. Based on real-time data and analysis results, medical resources and processes can be flexibly adjusted. Analysis models are continuously updated during training and verification to adapt to emerging patient characteristics and trends, enhancing the hospital's ability to respond to emergencies and improving the resilience of the medical system. At the same time, it also ensures that the hospital can continuously optimize services based on the latest medical knowledge and technology. Compared with existing technologies, by obtaining access to the hospital's electronic health record system, the basic information and medical history of the patient can be quickly and accurately read, reducing the possibility of manual input errors and improving the timeliness and reliability of data. By recording the patient's current symptoms and signs in real time, it ensures that the doctor obtains the latest patient status information, assists in formulating more reasonable treatment plans, and improves the accuracy of diagnosis. By building a deep learning analysis model to process historical medical record data, complex features are extracted and accurately evaluated, replacing simple statistical methods, providing more accurate disease assessment and prediction. By inputting patient data into a trained analysis model, the disease severity coefficient is automatically generated, reducing the subjectivity of manual scoring and improving the objectivity and consistency of the evaluation results. By establishing a strategy library and matching according to the severity coefficient of the disease, the most suitable hospitalization plan can be quickly found for the patient, which improves the response speed and the ability of personalized service, and optimizes the patient's admission experience. By recording the attributes of the nursing level beds, the bed utilization rate and management efficiency are improved, and the risk of information loss in traditional manual management is reduced. Automatic sorting and allocation of beds according to the severity coefficient ensures that patients with serious conditions receive appropriate care first, avoiding waste of resources and unreasonable allocation, and improving bed utilization efficiency.

[0023] Embodiment 2: In other aspects, this embodiment also provides an analysis model calculation formula. The analysis model expresses the disease severity coefficient through a multivariate linear regression strategy. The specific calculation formula is: ; In the formula, Represents the severity coefficient, which indicates the severity of the target patient's condition. The larger the value, the more serious the condition. represents the intercept term, which indicates the baseline severity coefficient when all eigenvalues ​​are zero, Represents the number of label features, such as personal information, medical history, symptoms and signs, Represents the coefficient of linear regression, indicating the characteristics The degree of influence on the severity coefficient S, represents the value of the i-th feature, such as the patient's age, the severity of past medical history, and the severity of current symptoms, Represents the number of features used to calculate nonlinear effects, represents the nonlinear regression coefficient, Representation characteristics The degree of influence of self-influence on the severity of the disease, represents the square of the jth feature, indicating the nonlinear effect of this feature on the severity of the disease. Represents the number of features and is used to calculate the interaction effects between features. Represents the interaction coefficient, indicating the characteristics With features The effect of the interaction term on the severity coefficient, Represents the product of the value of feature k and the value of feature l, indicating the interaction between feature k and feature l. represents the error term, which represents the partial or random error that the model cannot explain and is assumed to follow a normal distribution with a mean of zero; By constructing such a formula, the analysis model can effectively analyze the impact of various characteristics and the interaction between them, thereby providing a comprehensive severity coefficient. This coefficient can not only help doctors better understand the patient's condition, but also comprehensively consider special factors in various aspects, thereby providing a more accurate severity coefficient assessment, which helps hospitals make more scientific decisions in resource allocation and patient management. The analysis model calculation formula Captures the direct impact of each feature on the severity coefficient, indicating that the contribution of each feature to the severity coefficient is linear, through the square term of the feature Reflecting the increase in certain characteristic degrees, the impact on the condition increases. For example, serious underlying diseases may affect the condition in a nonlinear way. Analyze the interaction between characteristics, showing the effect of the combination of two specific characteristics on the severity coefficient, for example, the combination of age and a certain symptom may lead to a more severe disease than when the two characteristics are considered separately; As a preferred implementation in this embodiment, Figure 2 As shown, the working logic of the analysis model calculation formula is: Collect relevant information and label characteristics of target patients; Select several important features based on the availability and relevance of information; A feature matrix is ​​constructed based on the selected features, and the feature values ​​of each patient form a vector, for example, the patient’s age, number of underlying diseases, severity of current symptoms, and delay in seeking medical treatment; Use historical patient data to fit the model, that is, estimate the parameters of the model , and , through the fitting model, determine the influence of the characteristics on the severity coefficient of the disease; Using the fitted parameters and the new patient characteristic values, calculate the severity coefficient; Interpret the calculated severity coefficient and judge the severity of the patient's condition. The higher the severity coefficient, the more serious the condition. Make clinical decisions based on the results of the severity coefficient. Continuously monitor and evaluate the performance of analytical models, validate them with new patient data, and update analytical models.

[0024] Compared with existing technologies, the analysis model extracts the patient's personal information, medical history, symptoms and treatment plans, and can generate a quantitative disease severity coefficient. This objective quantitative indicator can more accurately reflect the patient's health status and provide a scientific basis for subsequent medical decision-making. In traditional medicine, the severity assessment of a disease often relies on the subjective judgment of doctors, which is easily affected by personal experience, emotions and other factors. However, deep learning algorithms can reduce human subjective errors and make the results more objective and reliable. The automated generation of disease severity coefficients can quickly provide doctors with health assessments of patients, reducing the time required for manual recording and analysis, allowing doctors to make decisions more quickly and take timely measures. It can update the severity coefficient in real time as the patient's symptoms, signs and treatment plans are adjusted. Through dynamic evaluation, doctors can timely understand the development and changes of the patient's condition and better adjust the treatment strategy. By integrating the patient's historical data, current condition and treatment plan, multi-dimensional features are extracted, and the data model formed has high information richness and accuracy, which can reflect more complex disease conditions; The generated disease severity coefficient can be used as an important indicator in clinical research and data analysis, helping hospitals evaluate the effectiveness of different treatment options and promoting continuous improvement in medical quality.

[0025] In summary, the present invention uses deep learning algorithms and analysis models to accurately evaluate the patient's condition. Through multi-dimensional analysis of the patient's basic information, medical history, symptoms and treatment plan, the model can generate a more accurate disease severity coefficient. Combined with the disease severity coefficient and real-time information of nursing beds, it automatically matches appropriate beds and nursing levels. By giving priority to allocating relevant resources to patients with more serious conditions, the hospital's bed resources are maximized, which helps to improve the overall nursing quality and efficiency, improve the accuracy of the hospital's initial triage, and provide personalized medical plans according to the patient's specific condition, thereby improving the treatment effect; Through automated data collection and processing, the time spent by medical staff on patient registration and information entry is reduced. Real-time recording of patients' symptoms and signs allows information to be quickly captured and analyzed, reducing patient waiting time, improving the response speed of medical services, and enabling doctors to make treatment decisions more quickly. Based on real-time data and analysis results, medical resources and processes can be flexibly adjusted. Analysis models are continuously updated during training and verification to adapt to emerging patient characteristics and trends, enhancing the hospital's ability to respond to emergencies and improving the resilience of the medical system. At the same time, it also ensures that the hospital can continuously optimize services based on the latest medical knowledge and technology. Through deep learning model training, the method of outputting the severity coefficient of the disease can integrate historical data and real-time data. The objectivity and accuracy are much higher than a single manual scoring system or empirical judgment. The automated coefficient generation mechanism can greatly reduce the workload of doctors, allowing them to focus on more complex medical judgments, thereby improving the overall efficiency and quality of medical services. This intelligent management planning achieves higher accuracy, efficiency and patient satisfaction through automation, deep learning technology and intelligent matching strategies, thereby optimizing the allocation and management of medical resources and improving the operation level of the hospital. Compared with existing technologies, this method can better adapt to rapidly changing medical needs and provide timely, scientific and personalized medical services.

[0026] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A pre-hospital registration method for intelligent management planning, characterized in that: The following steps are involved: Step 1: Obtain access to the hospital's electronic health record system to read the target patient's basic personal information and medical history; Step 2: Record the current personal symptoms and signs of the target patient in real time, and obtain the treatment plan currently formulated by the doctor for the target patient; Step 3: Use deep learning algorithms to build an analysis model, obtain historical medical record data through the hospital's electronic health record system, and extract label features of several patients' personal information, medical history, symptoms, signs, treatment plans, hospitalization time, and treatment effects, build a training sample set, and input it into the analysis model for training; Step 4: Obtain the trained analysis model, extract the label features of the current target patient's personal information, medical history, symptoms, and treatment plan, form a label feature set, input the target patient's label feature set into the analysis model, and output the target patient's disease severity coefficient; Step 5: Construct a strategy library of patient condition severity coefficients and hospitalization plans corresponding to different label feature combinations, and match the hospitalization plan required by the current target patient in the strategy library based on the target patient condition severity coefficient currently output; Step 6: Enter the attributes of all beds of different nursing levels in the current hospital; Step 7: According to the patient's condition severity coefficient and the matched hospitalization plan, the nursing level beds required by the current patient are sorted and allocated, with priority given to patients with serious conditions, and a bed allocation list is generated and submitted to the management end.

2. The pre-hospital registration method for intelligent management planning according to claim 1, characterized in that: The analysis model in step 3 represents the severity coefficient through a multiple linear regression strategy, and the calculation formula of the analysis model is: ； In the formula, Represents the severity coefficient of the disease, represents the intercept term, Represents the number of label features, represents the coefficient of linear regression, represents the value of the i-th feature, represents the number of features, represents the nonlinear regression coefficient, represents the square of the jth feature, represents the number of features, represents the interaction coefficient, represents the product of the value of feature k and the value of feature l, represents the error term.

3. The pre-hospital registration method for intelligent management planning according to claim 2, characterized in that: The analysis model calculation formula Captures the direct impact of each feature on the severity coefficient, indicating that the contribution of each feature to the severity coefficient is linear, through the square term of the feature Reflecting the increase in certain characteristic degrees, the impact on the disease is enhanced, through The interaction between features was analyzed to show the effect of the combination of two specific features on the severity coefficient.

4. The pre-hospital registration method for intelligent management planning according to claim 2, characterized in that: The working logic of the analysis model calculation formula is: Collect relevant information and label characteristics of target patients; Select several important features based on the availability and relevance of information; A feature matrix is ​​constructed based on the selected features, and the feature values ​​of each patient form a vector; Use historical patient data to fit the model, and determine the degree of influence of the characteristics on the severity coefficient of the disease by fitting the model; Using the fitted parameters and the new patient characteristic values, calculate the severity coefficient; Interpret the calculated severity coefficient and judge the severity of the patient's condition. The higher the severity coefficient, the more serious the condition. Make clinical decisions based on the results of the severity coefficient. Continuously monitor and evaluate the performance of analytical models, validate them with new patient data, and update analytical models.

5. The pre-hospital registration method for intelligent management planning according to claim 1, characterized in that: The process of matching the hospitalization plan required by the current target patient in the strategy library in step 5 is as follows: A hospitalization strategy library containing detailed information of a plurality of hospitalization plans is predefined, and each hospitalization plan corresponds to a different severity coefficient interval and patient characteristics; Using the target patient's disease severity coefficient as a key indicator, search the hospitalization plans in the strategy library, and use conditional judgment or query statements to determine which hospitalization plans have the disease severity coefficient within the condition range; Output the hospitalization plan whose severity coefficient meets the condition range.

6. The pre-hospital registration method for intelligent management planning according to claim 1, characterized in that: In step 3, the training sample set is divided into 70% training set and 30% validation set, and the ratio of the training set to the validation set is adjusted by receiving external instructions.

7. The pre-hospital registration method for intelligent management planning according to claim 1, characterized in that: In step 5, if the target patient's condition severity coefficient currently output does not match the hospitalization plan required by the current target patient in the current policy library, the index is terminated, a supplementary report of the associated data of the condition severity coefficient is generated, and submitted to the management end.

8. The pre-hospital registration method for intelligent management planning according to claim 1, characterized in that: The bed attributes in step 6 include: bed status, number of beds, nursing level and regional location.

9. The pre-hospital registration method for intelligent management planning according to claim 1, characterized in that: The working logic of the sorting and allocation in step 7 is: traverse the sorted patient list, check the available beds according to the required nursing level of the current patient, and allocate corresponding beds to eligible patients. If the corresponding bed is full, skip the bed and continue to check the next bed. Once a bed is allocated to the patient, update the bed as occupied and update the available bed information. If a bed is not allocated to the patient, send a prompt message to the management end.

10. The pre-hospital registration method for intelligent management planning according to claim 1, characterized in that: The bed allocation list in step 7 includes: patient identity information, allocated bed information, nursing level, hospitalization time information and disease severity coefficient.

Images