Intelligent nursing method for severe hyperpyrexia patient

Abstract

The present application relates to the technical field of intelligent nursing, and specifically discloses an intelligent nursing method for a patient with severe high fever, which comprises the following steps: acquiring vital sign change parameters of the patient with severe high fever in a monitoring period and disease condition change characteristics of the patient with severe high fever in an effective nursing period, inputting the parameters and the characteristics into a recurrent neural network model to train and output a nursing grade result; the nursing grade is divided into first-class nursing, second-class nursing and third-class nursing in order from high to low according to the severity of nursing intervention; acquiring monitoring safety risk prompt information according to the nursing grade result; calling nursing grade parameters in a preset nursing grading system according to the monitoring safety risk prompt information; inputting the called nursing grade parameters into the trained recurrent neural network model to output an adjustment coefficient corresponding to current nursing grade data, and judging the size of the current adjustment coefficient to carry out nursing work; the present application improves the accuracy of intelligent nursing grading, achieves the purpose of high-quality and safe nursing for patients with severe high fever, and improves the clinical value.

Description

Technical Field

[0001] This invention relates to the field of intelligent nursing technology, specifically to an intelligent nursing method for patients with severe high fever. Background Technology

[0002] The treatment and management of patients with severe high fever is a complex process that requires comprehensive consideration of the patient's specific condition and underlying cause. When treating patients with severe high fever, the medical team needs to closely monitor changes in the patient's condition and adjust the treatment plan based on the latest clinical guidelines and research findings. Meanwhile, integrating traditional Chinese and Western medicine can provide a more comprehensive treatment strategy to improve treatment outcomes and patient survival rates.

[0003] Existing intelligent early warning systems for patient safety can monitor patients' nursing status to a certain extent by real-time monitoring of their vital signs, current monitoring procedures, and monitoring status. However, with the surge in infectious disease patients in recent years, the corresponding increase in the number of nursing staff, nursing wards, and monitoring procedures requires the deployment of more nursing personnel to cope with the situation. In addition, it is necessary to increase the number of monitoring terminals and the acquisition of real-time monitoring data by relevant management personnel. Due to the limited area of ​​treatment areas and the diverse disease types of patients with severe high fever, isolation and monitoring measures are also required. A large number of monitoring personnel is actually not conducive to the treatment and recovery process of patients with severe high fever.

[0004] While an intelligent and systematic nursing grading system based on the Internet of Things has been designed and applied in nursing work, the increasing number of patients with severe high fever makes the solution of directly increasing the number of nursing staff no longer suitable for the nursing process of a large number of patients with severe high fever due to the limited number of nursing staff. Furthermore, the quality and efficiency of existing grading nursing cannot be directly reflected through data evaluation. It is necessary to analyze the current nursing staff's detection rate of potential complications, postoperative complications, and adverse events of current patients. However, for patients with severe high fever, the urgency of treatment and the severity of life safety risks are significant. Using questionnaire-based statistical methods affects the actual treatment and rehabilitation efficiency of patients and has low clinical value. Summary of the Invention

[0005] The purpose of this invention is to provide an intelligent nursing method for patients with severe high fever, and to solve the following technical problems:

[0006] How can we improve the accuracy of intelligent nursing triage to achieve high-quality and safe nursing care for critically ill patients with high fever, while also enhancing clinical value?

[0007] The objective of this invention can be achieved through the following technical solutions:

[0008] Intelligent nursing methods for patients with severe high fever include:

[0009] S1. Obtain the vital sign change parameters of critically ill patients with high fever during the monitoring period and the condition change characteristics of critically ill patients with high fever during the effective nursing cycle. Input the data into the recurrent neural network model for training and output the nursing level results.

[0010] Changes in vital signs parameters include: changes in body temperature, changes in heart rate, and changes in blood pressure;

[0011] The characteristics of changes in the condition include: improvement, deterioration, and stabilization.

[0012] Nursing care levels are classified into Level 1, Level 2, and Level 3 nursing care, from highest to lowest, based on the severity of the nursing intervention.

[0013] S2. Obtain monitoring safety risk warning information based on the nursing level results;

[0014] S3. Retrieve nursing level parameters from the preset nursing grading system based on the monitoring safety risk warning information;

[0015] S4. Input the retrieved nursing level parameters into the currently trained recurrent neural network model, output the adjustment coefficient corresponding to the current nursing level data, and determine the magnitude of the current adjustment coefficient to carry out nursing work.

[0016] Preferably, the method for obtaining nursing grade data in step S1 is as follows:

[0017] Determine the set of vital sign changes in patients with severe high fever during the monitoring period. ;

[0018] Determine the set of characteristic parameters of disease changes in all patients with severe high fever during all effective nursing cycles. , The number of effective nursing cycles;

[0019] Obtain the set of characteristic parameters of disease condition changes within each effective nursing cycle. All elements The set of vital sign change parameters corresponding to the same time point Combine them to form a new set. and set ∈ ; This is the current effective nursing cycle; This represents the total number of time points within an effective nursing cycle.

[0020] New set All elements are used as the training set and input into the recurrent neural network model for training. The effective nursing value corresponding to each element is obtained, and the current effective nursing value is divided into large to small according to the preset effective nursing value threshold range and the nursing level result is output.

[0021] Preferably, the effective nursing value is calculated as follows:

[0022] Through formula Calculate the real-time effective nursing value ;

[0023] in, At the current time, and ∈ ; , All are preset weighting coefficients, and , All are greater than 0; Preset nursing functions; These are the vital signs change parameters at the current time point. These are standard vital sign change parameters at the same time point. This is the preset deviation value for the vital signs change parameters at the current time point; This is a function representing the transformation of the patient's condition. These are characteristic changes in the patient's condition; These are the characteristic parameters of the disease's changes at the current point in time. These are the standard characteristic parameters of disease progression at the current point in time. For the first The impact coefficient of an effective nursing cycle.

[0024] Preferably, the nursing grade results include:

[0025] Current effective nursing value Compared with the preset effective nursing value threshold range Compare:

[0026] like < If the current effective nursing quality is good and little nursing intervention is required, the nursing level is set as Level 3 nursing.

[0027] like ≤ ≤ If the current effective nursing care is deemed to be average and more nursing intervention is required, the nursing level is set as Level II nursing care.

[0028] like > If the current effective nursing care is poor, a lot of nursing intervention is needed, and the nursing level is set as Level 1 nursing care.

[0029] Preferably, the monitoring safety risk warning information includes:

[0030] The nursing grade signal for each critically ill patient with high fever is determined based on the nursing grade results.

[0031] The nursing staff information is matched to the corresponding nursing staff level based on the nursing level signal; the nursing staff information includes the nursing staff level and nursing staff experience parameters;

[0032] Early warning signals are generated based on the results of the Level 1 nursing care assessment to monitor changes in the nursing status of critically ill patients with high fever at Level 1 nursing care.

[0033] Preferably, in step S3:

[0034] Input the nursing grade signals of all critically ill patients with high fever and the corresponding nursing staff information into the analysis model of the preset nursing grade system, and output the nursing grade parameters;

[0035] It also provides adjustment information for critically ill patients with high fever at the first level of nursing care based on the early warning signal feedback, and generates an adjustment table.

[0036] Preferably, in step S4:

[0037] Through formula Calculate the real-time adjustment coefficient ;

[0038] in, A preset time period within the effective nursing cycle time point; For the first Real-time nursing level parameters within the cumulative time period of each effective nursing cycle. For the first Preset standard nursing level parameters for each effective nursing cycle. For the first Deviation values ​​of preset nursing level parameters for each effective nursing cycle; This is a conversion function for nursing grade parameters; For the first Real-time effective nursing value for each effective nursing cycle.

[0039] Preferably, the adjustment coefficient With preset adjustment coefficient threshold Compare:

[0040] like ≥ If the current adjustment coefficient is too large, the current nursing level will be adjusted, and nursing work will be carried out accordingly based on the adjustment result.

[0041] like < If the current adjustment coefficient is too small, the current nursing level should be maintained, and the current nursing care should continue.

[0042] The beneficial effects of this invention are as follows: This invention obtains vital sign change parameters of critically ill patients with high fever during the monitoring period and condition change characteristics of critically ill patients with high fever during the effective nursing cycle, inputs them into a recurrent neural network model for training, and outputs nursing level results; it reflects the nursing status of critically ill patients with high fever through monitoring safety risk warning information, and accurately matches corresponding nursing personnel according to the nursing level signal to form a nursing and safety monitoring process for critically ill patients with high fever of different levels. Based on the monitoring safety risk warning information, it retrieves nursing level parameters from a preset nursing grading system; the retrieved nursing level parameters are input into the currently trained recurrent neural network model to output the adjustment coefficient corresponding to the current nursing level data, and the magnitude of the current adjustment coefficient is determined to carry out nursing work.

[0043] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0044] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 This is a flowchart illustrating the steps of the intelligent nursing method for critically ill patients with high fever according to the present invention. Detailed Implementation

[0046] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] While a smart and data-driven nursing grading system based on the Internet of Things can be built to ensure its application in nursing work, the existing method of directly increasing the number of critically ill patients with high fever is no longer directly applicable to the actual situation. Furthermore, the quality and efficiency of the existing grading nursing cannot be directly reflected through data-driven evaluation results. Instead, it is necessary to analyze the current nursing staff's detection rate of potential complications, postoperative complications, and adverse events. However, for critically ill patients with high fever, the urgency of treatment and the severity of life-threatening risks are significant. Using questionnaire-based statistical methods affects the actual treatment and rehabilitation efficiency of patients, resulting in low clinical value.

[0048] Please see Figure 1As shown, in order to solve the above-mentioned technical problems, the present invention provides an intelligent nursing method for patients with severe high fever, the method comprising:

[0049] S1. Obtain the vital sign change parameters of critically ill patients with high fever during the monitoring period and the condition change characteristics of critically ill patients with high fever during the effective nursing cycle. Input the data into the recurrent neural network model for training and output the nursing level results.

[0050] Changes in vital signs parameters include: changes in body temperature, changes in heart rate, and changes in blood pressure;

[0051] The characteristics of changes in the condition include: improvement, deterioration, and stabilization.

[0052] Nursing care levels are classified into Level 1, Level 2, and Level 3 nursing care, from highest to lowest, based on the severity of the nursing intervention.

[0053] S2. Obtain monitoring safety risk warning information based on the nursing level results;

[0054] S3. Retrieve nursing level parameters from the preset nursing grading system based on the monitoring safety risk warning information;

[0055] S4. Input the retrieved nursing level parameters into the currently trained recurrent neural network model, output the adjustment coefficient corresponding to the current nursing level data, and determine the magnitude of the current adjustment coefficient to carry out nursing work.

[0056] In the above technical solution, this embodiment obtains the nursing level by analyzing existing disease state monitoring results, provides monitoring safety risk warnings, determines the rationality of the nursing level parameters based on the monitoring safety risk warning information, and finally achieves further precision in nursing-related data by adjusting the nursing level; the specific steps are as follows:

[0057] First, step S1 involves inputting the vital sign change parameters of critically ill patients with high fever during the monitoring period and the condition change characteristics of critically ill patients with high fever within the effective nursing cycle into a recurrent neural network model for training and outputting nursing level results. Among them, the vital sign change parameters analyzed based on medical physiological knowledge include: body temperature change value, heart rate change value, and blood pressure change value; the condition change characteristics are set as: condition improvement, condition deterioration, and condition stabilization; and the nursing level is divided into Level 1 nursing, Level 2 nursing, and Level 3 nursing according to the severity of nursing intervention from high to low. Unlike the general method of directly assigning nursing staff to patients, this design judges the patient's nursing needs from the perspective of nursing intervention and then provides a more appropriate nursing approach.

[0058] The recurrent neural network model is used to train on the vital sign changes of critically ill patients with high fever during the monitoring period and the sequence data of their condition changes within the effective nursing cycle. It determines the relationship between patients' self-care ability and nursing needs to obtain nursing grade levels by dividing historical data with labeled nursing grade results into training, validation, and test sets. For example, based on past medical records of critically ill patients with high fever, their vital sign changes, condition changes, and corresponding nursing grade results are determined, and then divided into training, validation, and test sets according to a certain ratio (e.g., 7:2:1). During training, a loss function (e.g., cross-entropy loss function) is used to measure the difference between the model's output and the actual nursing grade results. The model's weights are continuously adjusted using the backpropagation algorithm to minimize the loss function. During training, the loss and accuracy of the validation set are monitored to prevent overfitting. If the loss on the validation set starts to increase or the accuracy stops improving, it indicates that the model may be overfitting, requiring adjustments to the model structure or training parameters (e.g., reducing the learning rate, increasing the regularization term).

[0059] Specifically, the method for obtaining nursing grade data in step S1 is as follows:

[0060] Determine the set of vital sign changes in patients with severe high fever during the monitoring period. ;

[0061] Determine the set of characteristic parameters of disease changes in all patients with severe high fever during all effective nursing cycles. , The number of effective nursing cycles;

[0062] Obtain the set of characteristic parameters of disease condition changes within each effective nursing cycle. All elements The set of vital sign change parameters corresponding to the same time point Combine them to form a new set. and set ∈ ; This is the current effective nursing cycle; This represents the total number of time points within an effective nursing cycle.

[0063] New set All elements are used as the training set and input into the recurrent neural network model for training. The effective nursing value corresponding to each element is obtained, and the current effective nursing value is divided into large to small according to the preset effective nursing value threshold range and the nursing level result is output.

[0064] In the above technical solution, this embodiment achieves a matching reference for the nursing needs of critically ill patients with high fever by obtaining the nursing level. Specifically, it first determines the set of vital sign change parameters of critically ill patients with high fever during the monitoring period. Based on the combined analysis of changes in body temperature, heart rate, and blood pressure, a set of characteristic parameters of disease changes in all patients with severe high fever within all effective nursing cycles was determined. By statistically combining data on improvement, deterioration, and stabilization of patient conditions, and then calculating and analyzing these data, characteristic parameters of patient condition changes for each effective nursing cycle are obtained. Furthermore, these two sets of information are recombined through time matching to obtain a set of characteristic parameters of patient condition changes within each effective nursing cycle. All elements The set of vital sign change parameters corresponding to the same time point Combine the elements in the set to form a new set. Finally, the new set All elements are used as the training set and input into the recurrent neural network model for training. The effective nursing value corresponding to each element is obtained, and the current effective nursing value is divided into large to small according to the preset effective nursing value threshold range and the nursing level result is output.

[0065] The effective nursing cycle is relative to the information data on changes in the patient's condition. As long as the patient's treatment shows improvement or slows down deterioration within the time range of historical medical staff intervention, it can be called an effective nursing cycle. Such an effective nursing cycle is usually phased, and due to the phased nature of the hospital's treatment plan and the time span of different disease treatment courses during the treatment or rehabilitation process, there may also be multiple effective nursing cycles.

[0066] As one embodiment of the present invention, this embodiment calculates the effective nursing value, and the specific calculation method is as follows:

[0067] Through formula Calculate the real-time effective nursing value ;

[0068] in, At the current time, and ∈ ; , All are preset weighting coefficients, and , All are greater than 0; Preset nursing functions; These are the vital signs change parameters at the current time point. These are standard vital sign change parameters at the same time point. This is the preset deviation value for the vital signs change parameters at the current time point; This is a function representing the transformation of the patient's condition. These are characteristic changes in the patient's condition; These are the characteristic parameters of the disease's changes at the current point in time. These are the standard characteristic parameters of disease progression at the current point in time. For the first The impact coefficient of an effective nursing cycle.

[0069] In the above technical solution, the effective nursing value can be used to directly determine the nursing quality corresponding to the current element, namely, the nursing needs of critically ill patients with high fever; by analyzing the parameters of changes in vital signs and the parameters of changes in the condition as the main factors, the effective nursing value can be determined, thus achieving accurate confirmation of the effective nursing value at the current time point.

[0070] It needs to be explained that the preset weighting coefficients , All settings are selected based on empirical data. Specific settings can be quantitatively or qualitatively determined by assessing the proportion of the impact of changes in vital signs and the characteristics of changes in the patient's condition on the overall effective nursing outcomes within the total nursing needs; [Patient change transformation function] This data was obtained by fitting test data of characteristic parameters of changes in the patient's condition under normal nursing care.

[0071] Predefined functions typically refer to rules or logic predefined in certain systems or algorithms to guide the allocation process of resources, data, or tasks; the predefined nursing function in this design... It is a customized adjustment function based on historical data, tailored to the specific circumstances and changes in the patient's condition, and ensures that the result calculated by the current formula is within a reasonable range of effective nursing values.

[0072] , The average or average variance of the data on the impact of changes in vital signs and changes in patient condition under normal effective nursing needs, respectively, is determined by the data obtained through machine pre-simulation of the impact of changes in vital signs and changes in patient condition characteristics under normal effective nursing needs. The settings were selected based on empirical data and will not be detailed here; the influence coefficient of the effective nursing cycle. The settings are selected based on the comprehensive nursing status of patients with severe high fever in different effective nursing cycles, and will not be described in detail here.

[0073] As one embodiment of the present invention, the nursing grade result includes:

[0074] Current effective nursing value Compared with the preset effective nursing value threshold range Compare:

[0075] like < If the current effective nursing quality is good and little nursing intervention is required, the nursing level is set as Level 3 nursing.

[0076] like ≤ ≤ If the current effective nursing care is deemed to be average and more nursing intervention is required, the nursing level is set as Level II nursing care.

[0077] like > If the current effective nursing care is poor, a lot of nursing intervention is needed, and the nursing level is set as Level 1 nursing care.

[0078] In the above technical solution, the nursing level result in this embodiment is determined by the effective nursing value. The size is determined by comparing it with a preset effective nursing value threshold range. Compare the sizes and determine if... < If the current effective nursing quality is good and requires little nursing intervention, the nursing level is set as Level 3 nursing; if ≤ ≤ If the current effective nursing care is deemed insufficient and requires more nursing intervention, the nursing level is determined to be Level II; if > If the current effective nursing care is deemed insufficient, requiring significant nursing intervention, the nursing level is determined to be Level 1 nursing care; this achieves precise determination of the nursing level. It is important to note that this is based on real-time effective nursing care values ​​within the effective treatment cycle or a preset time period. The patient's final nursing care level is determined based on a comprehensive assessment.

[0079] Then, step S2 obtains monitoring safety risk warning information based on the nursing level results; the monitoring safety risk warning information reflects the nursing status of critically ill patients with high fever, and accurately matches the corresponding nursing staff according to the nursing level signal to form a nursing and safety monitoring process for critically ill patients with high fever of different levels.

[0080] As one embodiment of the present invention, the monitoring safety risk warning information includes:

[0081] The nursing grade signal for each critically ill patient with high fever is determined based on the nursing grade results.

[0082] The nursing staff information is matched to the corresponding nursing staff level based on the nursing level signal; the nursing staff information includes the nursing staff level and nursing staff experience parameters;

[0083] Early warning signals are generated based on the results of the Level 1 nursing care assessment to monitor changes in the nursing status of critically ill patients with high fever at Level 1 nursing care.

[0084] In the above technical solution, specifically regarding the process of obtaining monitoring safety risk alerts based on the nursing level assessment results, the monitoring safety risk alert information mainly includes: confirming the nursing level signal for each critically ill patient with high fever based on the nursing level results; then, confirming the nursing staff information matching the corresponding level based on the nursing level signal; the nursing staff information includes the nursing staff level and nursing staff experience parameters; and also generating an early warning signal based on the Level 1 nursing level results to monitor changes in the nursing status of critically ill patients with high fever at Level 1 nursing level.

[0085] Next, step S3 retrieves the nursing level parameters from the preset nursing grading system based on the monitoring safety risk warning information; specifically, it inputs the nursing level signals of all critically ill patients with high fever and the corresponding nursing staff information into the analysis model of the preset nursing grading system, outputs the nursing level parameters, and provides adjustment information for critically ill patients with high fever at the first-level nursing level based on the warning signal, and generates an adjustment table.

[0086] In the above technical solution, the analysis and monitoring of safety risk warning information in the previous step includes prompting the patient's nursing level, assigning patient care personnel, and training the current displayed information through the analysis model of the existing nursing grading system to achieve the result of nursing level parameterization, outputting the current nursing level parameters, and also providing feedback on adjustment information for critically ill patients with high fever at the first level of nursing care based on the warning signal, and generating an adjustment table.

[0087] Finally, step S4 involves inputting the retrieved nursing level parameters into the currently trained recurrent neural network model to output the adjustment coefficients corresponding to the current nursing level data. Combining the output of the nursing level parameters, the nursing level data in the pre-trained recurrent neural network model is readjusted, and its corresponding adjustment coefficients are generated. Based on the magnitude of the adjustment coefficients, corresponding nursing work is carried out.

[0088] In the above technical solution, the effective nursing values ​​corresponding to the pre-acquired nursing level data are analyzed by using the input nursing level parameters, and the adjustment coefficient is output to update and adjust the nursing level data in a timely manner.

[0089] As one embodiment of the present invention, specifically, the method for obtaining the adjustment coefficient in step S4 is as follows:

[0090] Through formula Calculate the real-time adjustment coefficient ;

[0091] in, A preset time period within the effective nursing cycle time point; For the first Real-time nursing level parameters within the cumulative time period of each effective nursing cycle. For the first Preset standard nursing level parameters for each effective nursing cycle. For the first Deviation values ​​of preset nursing level parameters for each effective nursing cycle; This is a conversion function for nursing grade parameters; For the first Real-time effective nursing value for each effective nursing cycle.

[0092] In the above technical solution, the relationship between the preset nursing level parameters of the effective nursing cycle and the implemented effective nursing value is used to obtain the real-time adjustment coefficient. Mainly through transformation functions The transformation of the reference values ​​of the current preset nursing level parameters, where the transformation function... The real-time adjustment coefficient is obtained by fitting nursing test data within the current effective nursing cycle cumulative time period, and by substituting the effective nursing value under the current state into the calculation.

[0093] It should be noted that the preset time period Based on empirical data, it can be adaptively adjusted according to the specific clinical monitoring needs; These are standard values ​​calculated based on changes in vital signs and characteristics of the patient's condition under the current normal nursing cycle. The settings were selected based on empirical data and will not be detailed here.

[0094] As one embodiment of the present invention, the adjustment coefficient With preset adjustment coefficient threshold Compare:

[0095] like ≥ If the current adjustment coefficient is too large, the current nursing level will be adjusted, and nursing work will be carried out accordingly based on the adjustment result.

[0096] like < If the current adjustment coefficient is too small, the current nursing level should be maintained, and the current nursing care should continue.

[0097] In the above technical solution, this embodiment further determines the adjustment information of the nursing level by analyzing the magnitude of the adjustment coefficient, and adjusts the nursing work according to the adjustment direction.

[0098] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments for apparatus, devices, and non-volatile computer storage media are relatively simple in description because they are fundamentally similar to the method embodiments; relevant parts can be referred to the descriptions of the method embodiments.

[0099] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0100] The above description is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined in the claims, they should all fall within the protection scope of the present invention.

Claims

1. An intelligent nursing system for critically ill patients with high fever, characterized in that: The steps by which the system performs the nursing method include: S1. Obtain the vital sign change parameters of critically ill patients with high fever during the monitoring period and the condition change characteristics of critically ill patients with high fever during the effective nursing cycle. Input the data into the recurrent neural network model for training and output the nursing level results. The vital signs parameters include: changes in body temperature, changes in heart rate, and changes in blood pressure; The characteristics of the disease changes include: improvement, deterioration, and stabilization. The nursing care level results are divided into Level 1, Level 2, and Level 3 nursing care in descending order of the severity of the nursing intervention. The method for obtaining nursing grade data in step S1 is as follows: Determine the set of vital sign changes in patients with severe high fever during the monitoring period. ; Determine the set of characteristic parameters of disease changes in all patients with severe high fever during all effective nursing cycles. , The number of effective nursing cycles; Obtain the set of characteristic parameters of disease condition changes within each effective nursing cycle. All elements The set of vital sign change parameters corresponding to the same time point Combine them to form a new set. and set ∈ ; This is the current effective nursing cycle; This represents the total number of time points within an effective nursing cycle. New set All elements are used as the training set and input into the recurrent neural network model for training. The effective nursing value corresponding to each element is obtained, and the current effective nursing value is divided into large to small according to the preset effective nursing value threshold range and the nursing level result is output. The effective nursing value is calculated as follows: Through formula Calculate the real-time effective nursing value ; in, At the current time, and ∈ ; , All are preset weighting coefficients, and , All are greater than 0; Preset nursing functions; These are the vital signs change parameters at the current time point. These are standard vital sign change parameters at the same time point. This is the preset deviation value for the vital signs change parameters at the current time point; This is a function representing the transformation of the patient's condition. These are characteristic changes in the patient's condition; These are the characteristic parameters of the disease's changes at the current point in time. These are the standard characteristic parameters of disease progression at the current point in time. For the first The influence coefficient of an effective nursing cycle; S2. Obtain monitoring safety risk warning information based on the nursing level results; The monitoring and safety risk warning information includes: The nursing grade signal for each critically ill patient with high fever is determined based on the nursing grade results. The nursing staff information is matched to the corresponding nursing staff level based on the nursing level signal; the nursing staff information includes the nursing staff level and nursing staff experience parameters. Early warning signals are generated based on the Level 1 nursing care level to monitor changes in the nursing status of critically ill patients with high fever at Level 1 nursing care level. S3. Retrieve nursing level parameters from the preset nursing grading system based on the monitoring safety risk warning information; In step S3: Input the nursing grade signals of all critically ill patients with high fever and the corresponding nursing staff information into the analysis model of the preset nursing grade system, and output the nursing grade parameters; And based on the feedback of early warning signals, it provides adjustment information for critically ill patients with high fever at the first level of nursing care, and generates an adjustment table; S4. Input the retrieved nursing grade parameters into the currently trained recurrent neural network model to output the adjustment coefficient of the current nursing grade result, so as to carry out nursing work according to the current adjustment coefficient. In step S4: Through formula Calculate the adjustment coefficient ; in, A preset time period within the effective nursing cycle time point; For the first Real-time nursing level parameters within the cumulative time period of each effective nursing cycle. For the first Preset standard nursing level parameters for each effective nursing cycle. For the first Deviation values ​​of preset nursing level parameters for each effective nursing cycle; This is a conversion function for nursing grade parameters; For the first Real-time effective nursing value for each effective nursing cycle.

2. The intelligent nursing system for critically ill patients with high fever according to claim 1, characterized in that, The nursing care level results include: Current effective nursing value Compared with the preset effective nursing value threshold range Compare: like < If the current effective nursing care is deemed excellent and requires little nursing intervention, the nursing level is determined to be Level 3 nursing care. like ≤ ≤ If the current effective nursing quality is deemed to be average and requires more nursing intervention, the nursing level is set as Level II nursing. like > If the current effective nursing quality is poor, more nursing intervention is needed, and the nursing level is set as Level 1 nursing.

3. The intelligent nursing system for critically ill patients with high fever according to claim 1, characterized in that, Adjust the coefficient With respect to the preset adjustment coefficient threshold range Compare: like ∈ If the current adjustment coefficient is normal, the current nursing level should be maintained. like > If the current adjustment coefficient is too high, the current nursing level will be lowered. like < If the current adjustment coefficient is too small, the current nursing level will be increased.

Images