An analysis system based on the data model of neurology nursing field

Abstract

The present invention relates to the field of medical nursing technology, and in particular to an analysis system based on a data model in the field of neurology nursing. The system includes a neurology nursing behavior feature analysis module, a neurology nursing risk assessment module, a nursing demand and feature change law analysis module, and a neurology nursing decision support module. It can obtain neurology patient behavior dimension data and neurology nursing record dimension data to perform condition nursing behavior feature analysis and neurology nursing risk assessment calculation to obtain a neurology condition nursing risk assessment value; obtain neurology patient nursing needs and perform feature change law mining and analysis, and at the same time perform nursing decision support analysis on the corresponding neurology patient nursing process, generate a neurology patient nursing decision support recommendation plan, and execute corresponding neurology patient nursing decision intervention measures. The present invention can significantly improve nursing efficiency.

Description

Technical Field

[0001] The present invention relates to the field of medical nursing technology, and in particular to an analysis system based on a data model in the field of neurology nursing. Background Art

[0002] In recent years, with the rapid development of medical big data and the application of artificial intelligence technology, intelligent analysis methods based on data models have gradually entered the field of neurology nursing. By real-time collection and analysis of multi-dimensional information such as patients' clinical data, vital signs, and medical history records, it can provide a scientific basis for nursing decisions and improve the quality and efficiency of nursing services. By building a multi-dimensional, dynamically updated data model, it can realize intelligent analysis of the nursing process and behavioral characteristics of neurology patients. This method integrates patients' medical records, vital signs, neurological function indicators, psychological assessment results and other information, and uses machine learning and data mining technology to generate personalized nursing plans. It can analyze patients' health status and nursing needs in real time and automatically adjust nursing strategies to achieve the best nursing effect. However, traditional nursing management methods usually rely on manual records, clinical experience and standardized processes, but these methods often fail to meet the precise needs of patients with complex neurological diseases at different stages of care. Summary of the invention

[0003] Based on this, it is necessary for the present invention to provide an analysis system based on a data model in the field of neurology nursing to solve at least one of the above technical problems.

[0004] To achieve the above objectives, an analysis system based on the data model of neurology nursing field includes the following modules:

[0005] The neurology nursing behavior characteristic analysis module is used to obtain the neurology patient behavior dimension data and the neurology nursing record dimension data, and use the preset neurology nursing field data model to perform condition nursing behavior characteristic analysis on the neurology patient behavior dimension data and the neurology nursing record dimension data to obtain the neurology condition nursing behavior characteristics, including the behavior activity distribution loss and symptom onset probability of the neurology condition in each nursing record period;

[0006] The neurology nursing risk assessment module is used to calculate the neurology nursing risk assessment based on the behavioral activity distribution loss and symptom onset probability of the neurology condition in each nursing record period to obtain the neurology condition nursing risk assessment value;

[0007] The nursing demand and characteristic change law analysis module is used to obtain the nursing demand of neurology patients, and based on the nursing demand of neurology patients, conduct characteristic change law mining and analysis on the behavior activity distribution loss and symptom onset probability of neurology conditions in each nursing record period, so as to obtain the nursing change law between neurology nursing demand and behavior activity distribution loss and symptom onset probability;

[0008] The neurology nursing decision support module is used to perform nursing decision support analysis on the corresponding neurology patient nursing process based on the nursing change rules between neurology nursing needs and behavioral activity distribution loss and symptom onset probability, as well as the neurology disease nursing risk assessment value, and generate neurology patient nursing decision support recommendation plans to implement corresponding neurology patient nursing decision intervention measures.

[0009] Furthermore, the neurology nursing behavior characteristic analysis module includes the following functions:

[0010] By designing and applying a behavioral self-assessment scale for neurology patients and conducting a quantitative assessment of patient behavior using the behavioral self-assessment scale, we can obtain behavioral dimension data of neurology patients, including physiological behavioral changes and psychological behavioral changes of neurology patients.

[0011] By obtaining the corresponding neurology nursing record dimension data from the neurology electronic health record system, including neurology nursing frequency, neurology nursing content and neurology nursing patient response;

[0012] Conduct time series analysis on the neurology patient behavior dimension data and the neurology nursing record dimension data to generate the neurology patient behavior dimension time series and the neurology nursing record dimension time series;

[0013] A recurrent neural network was used to construct a data model for neurology nursing, and the data model was used to perform dynamic distribution analysis on the time series of neurology patient behavior and neurology nursing record dimensions under different nursing record time segments, to obtain the dynamic distribution of neurology patient behavior and the dynamic distribution of neurology nursing patient response under each nursing record period;

[0014] Based on the dynamic distribution of neurology patients' behaviors and the dynamic distribution of neurology patients' responses in each nursing record period, the nursing behavior characteristics of neurology patients' nursing process in the corresponding nursing record period were analyzed to obtain the behavioral activity distribution loss and symptom onset probability of neurology disease in each nursing record period.

[0015] Furthermore, the analysis of the nursing behavior characteristics of the neurology patients in the corresponding nursing record period based on the dynamic distribution of the neurology patients' behaviors and the dynamic distribution of the neurology nursing patients' responses in each nursing record period includes:

[0016] The behavior activity spatiotemporal distribution matrix of neurology patients' behavior dynamic distribution in each nursing record period is analyzed to generate the behavior activity distribution matrix of neurology patients corresponding to each nursing record period, including the time distribution and spatial distribution of the patient's physiological and psychological behavior activities;

[0017] Fourier transform was used to analyze the behavioral activity fluctuation transformation of the distribution matrix of neurology patients' behavioral activities in each nursing record period, and the corresponding waveform of neurology patients' behavioral activities in each nursing record period was obtained;

[0018] The fluctuation characteristics of the waveforms of the behavior and activity changes of neurology patients corresponding to each nursing record period are analyzed to obtain the fluctuation characteristics of the behavior and activity of neurology patients corresponding to each nursing record period; the fluctuation index is quantitatively calculated based on the fluctuation characteristics of the behavior and activity of neurology patients corresponding to each nursing record period to obtain the fluctuation index of the behavior and activity of neurology patients corresponding to each nursing record period;

[0019] The corresponding spatiotemporal distribution gradient of the behavior activities of neurology patients is obtained through the behavior activity distribution matrix of neurology patients in each nursing record period, and the behavior distribution loss evaluation and analysis of the nursing process of neurology patients in the corresponding nursing record period is performed based on the corresponding neurology patient behavior activity fluctuation index and the spatiotemporal distribution gradient of the behavior activities of neurology patients in each nursing record period, so as to obtain the behavior activity distribution loss of neurology disease in each nursing record period;

[0020] Based on the dynamic distribution of patient responses in neurology nursing in each nursing record period, a quantitative analysis of the probability of symptom onset of neurological disease in the nursing process of neurology patients in the corresponding nursing record period was performed to obtain the probability of symptom onset of neurological disease in each nursing record period.

[0021] Furthermore, the fluctuation characteristics of the behavioral activities of neurology patients specifically include the distribution amplitude of the behavioral activities of neurology patients, the distribution frequency of the behavioral activities of neurology patients, and the distribution phase of the behavioral activities of neurology patients.

[0022] Furthermore, the behavior distribution loss evaluation analysis of the nursing process of neurology patients in the corresponding nursing record period based on the behavioral activity fluctuation index of neurology patients in each nursing record period and the spatiotemporal distribution gradient of the behavioral activity of neurology patients includes:

[0023] Obtain the ideal distribution target of neurology patients' behavioral activities corresponding to each nursing record period;

[0024] Based on the temporal and spatial distribution gradient of the behavior activities of neurology patients corresponding to each nursing record period, the behavior activity difference analysis of the corresponding ideal distribution target of the behavior activities of neurology patients was performed to obtain the degree of difference between the behavior activities of neurology patients corresponding to each nursing record period and the ideal distribution target;

[0025] Based on the behavioral activity fluctuation index of neurological patients corresponding to each nursing record period and the degree of difference between the behavioral activities of neurological patients and the ideal distribution target, a behavioral distribution loss evaluation and analysis of the nursing process of neurological patients in the corresponding nursing record period was performed to obtain the behavioral activity distribution loss of neurological conditions in each nursing record period.

[0026] Furthermore, the quantitative analysis of the probability of symptom onset of the neurology patient nursing process in the corresponding nursing record period based on the dynamic distribution of the neurology patient nursing response in each nursing record period includes:

[0027] The dynamic distribution of patient responses in neurology nursing at each nursing record period is analyzed to obtain the corresponding neurology patient response symptom evolution map at each nursing record period;

[0028] The fluctuation range of the symptom evolution of neurology patients corresponding to each nursing record period is analyzed to obtain the fluctuation range of the symptom reaction of neurology patients corresponding to each nursing record period;

[0029] Based on the fluctuation amplitude of the corresponding neurology patients' reaction symptoms in each nursing record period, the corresponding neurology patients' reaction symptom evolution change map is analyzed for symptom outbreak nodes, and the corresponding neurology patients' reaction symptom outbreak point set in each nursing record period is obtained;

[0030] Based on the corresponding neurological patients' symptom outbreak point set in each nursing record period, the corresponding neurological patients' symptom evolution change map is used to quantitatively analyze the probability of symptom onset of the disease, so as to obtain the symptom onset probability of neurological disease in each nursing record period.

[0031] Further, the symptom outbreak node analysis of the corresponding neurology patient reaction symptom evolution change map based on the corresponding neurology patient reaction symptom fluctuation amplitude in each nursing record period includes:

[0032] The corresponding fluctuation rate of symptom reaction of neurology patients and the peak value of symptom reaction of neurology patients are obtained through the fluctuation amplitude of symptom reaction of neurology patients in each nursing record period;

[0033] The corresponding resonance time points of the neurology patients' reaction symptoms are obtained through the evolution change maps of the neurology patients' reaction symptoms in each nursing record period, and the fluctuation peak correlation analysis of the corresponding neurology patients' reaction symptom fluctuation kurtosis values ​​in each nursing record period is performed based on the resonance time points of the neurology patients' reaction symptoms, so as to obtain the corresponding neurology patients' reaction symptom fluctuation peak correlation factors in each nursing record period;

[0034] Based on the corresponding neurology patients' reaction symptom fluctuation peak correlation factors in each nursing record period, the corresponding neurology patients' reaction symptom fluctuation change rate and neurology patients' reaction symptom fluctuation peak value are analyzed to obtain the corresponding neurology patients' reaction symptom fluctuation outbreak threshold in each nursing record period;

[0035] Based on the corresponding outbreak threshold of neurological patients' reaction symptoms fluctuation in each nursing record period, the reaction symptom change value in the corresponding neurological patients' reaction symptom evolution change map is marked and judged for symptom outbreak nodes to obtain the corresponding neurological patients' reaction symptom outbreak point set in each nursing record period.

[0036] Furthermore, the neurology nursing risk assessment calculation in the neurology nursing risk assessment module is calculated by a nursing risk assessment calculation formula; wherein the nursing risk assessment calculation formula is specifically:

[0037] ;

[0038] In the formula, is the neurological condition nursing risk assessment value at time, is the total number of nursing record periods, is the item index of the nursing record period, For the The integral time range corresponding to each nursing record period, is the time variable parameter, For neurological diseases The probability of symptom onset during the nursing record period, is the risk weight coefficient for symptom onset, For neurological diseases The distribution intensity of nursing behavior activities in each nursing record period, is the risk weight coefficient of activity distribution intensity, For neurological diseases The distribution loss of behavioral activities under each nursing record period, is the nursing behavior risk weight coefficient, It is the correction factor for the nursing risk assessment value of neurological conditions.

[0039] Furthermore, the nursing needs and characteristic change law analysis module includes the following functions:

[0040] Capture neurology patient care needs;

[0041] Based on the nursing needs of neurology patients, a temporal correlation analysis was conducted on the behavioral activity distribution loss and symptom onset probability of neurology conditions in each nursing record period, and the temporal correlation between neurology nursing needs and behavioral activity distribution loss and symptom onset probability was obtained;

[0042] Based on the temporal correlation between neurology nursing needs, behavioral activity distribution loss, and symptom onset probability, the characteristic change rules of behavioral activity distribution loss and symptom onset probability of neurology conditions in each nursing record period were mined and analyzed to obtain the nursing change rules between neurology nursing needs, behavioral activity distribution loss, and symptom onset probability.

[0043] Furthermore, the neurology nursing decision support module includes the following functions:

[0044] Based on the nursing risk assessment value of the neurology condition, a nursing risk comparison and judgment is made on the corresponding neurology patient nursing process. If the nursing risk assessment value of the neurology condition is less than the preset nursing risk threshold, it is considered that there is no nursing risk event in the corresponding neurology patient nursing process; if the nursing risk assessment value of the neurology condition is greater than or equal to the preset nursing risk threshold, it is considered that there is a nursing risk event in the corresponding neurology patient nursing process;

[0045] Based on the nursing change rules between neurology nursing needs and behavioral activity distribution loss and symptom onset probability, a nursing decision support analysis is conducted on the nursing process of neurology patients with corresponding nursing risk events, and a nursing decision support recommendation plan for neurology patients is generated to implement corresponding nursing decision intervention measures for neurology patients, including comprehensively considering the nursing needs, disease risks and behavioral activity ability loss of neurology patients to recommend corresponding personalized nursing plans, optimize nursing plans and adjust nursing intervention measures.

[0046] Beneficial effects of the present invention:

[0047] The analysis system based on the data model of neurology nursing field proposed in the present invention is generally composed of a neurology nursing behavior characteristic analysis module, a neurology nursing risk assessment module, a nursing demand and characteristic change law analysis module and a neurology nursing decision support module. Compared with the prior art, the beneficial effect of the present application is that a structured data set is established by comprehensively collecting the behavioral dimension data and nursing record dimension data of neurology patients. The behavioral dimension data includes information on the patient's activity ability, cognitive status, emotional state, degree of pain, etc., while the nursing record dimension data involves various intervention measures in the nursing process, nursing observation data and patient reactions. The collection and integration of these data lays a solid foundation for subsequent analysis. By applying the preset data model for neurology nursing, in-depth analysis of the patient's behavioral dimension data and nursing record dimension data can reveal the patient's condition changes in different nursing record periods, the effect of nursing intervention and its correlation with behavioral activities. For example, the distribution loss of behavioral activities in each nursing period reflects the relationship between nursing intervention and patient behavior changes, while the symptom onset probability reveals the risk of recurrence of patient symptoms at different time points. These analysis results can provide nursing staff with specific disease dynamics and nursing response predictions, help formulate personalized nursing plans, thereby optimizing the patient's nursing process and improving the quality of nursing. Secondly, by calculating the nursing risk assessment based on the distribution loss of behavioral activities and the probability of symptom onset, the core purpose of this step is to provide a quantitative risk assessment model for nursing work by comprehensively analyzing the complexity of the patient's condition and the probability of symptom onset, helping nursing staff to predict the development trend of the patient's condition and potential nursing risks. The calculation of the nursing risk assessment value mainly depends on the patient's behavioral characteristic data and disease indicators, combined with the feedback of nursing intervention measures. High-risk patients often show significant loss of behavioral activities or a high probability of symptom recurrence. By calculating the nursing risk assessment value, high-risk situations can be identified in time during the patient's care process, thereby formulating accurate personalized nursing plans for subsequent processing processes, avoiding unnecessary nursing errors or delays, and ensuring that patients receive timely and accurate care.Then, by analyzing the regular changes between the patient's nursing needs and the loss of behavioral activity distribution and the probability of symptom onset, and by deeply mining the correlation between nursing needs and these data points, caregivers can have a clearer understanding of the patient's individualized nursing needs. For example, if a patient shows a significant decline in behavioral activities during a specific nursing period, or a high frequency of certain symptoms, it is closely related to their specific nursing needs, such as the need for more support, nursing adjustments or psychological counseling. This regular mining not only helps caregivers provide more accurate nursing services, but also helps to achieve real-time adjustments in the patient's care process. As nursing needs change, nursing intervention measures need to be continuously optimized to meet the different needs of patients. For the nursing team, accurately identifying these changing patterns can enhance the flexibility and timeliness of intervention, avoid worsening of the patient's condition or decline in nursing quality due to delayed or improper care, and thus minimize the risks caused by insufficient care or untimely intervention. Finally, by integrating nursing needs with behavioral activity distribution loss, the changing rules of symptom onset probability, and nursing risk assessment values ​​for nursing decision support analysis, a comprehensive decision-making framework can be provided for the nursing process of neurology patients. Through this analysis, the nursing team can develop more accurate and personalized nursing intervention plans to respond to the changing needs of patients in different nursing periods. This can provide real-time feedback on changes in the patient's condition and provide reasonable nursing decision-making suggestions based on the data. For example, for high-risk patients, the system can automatically recommend enhanced nursing intervention measures or adjust nursing strategies to prevent a sharp deterioration of the condition. For low-risk patients, excessive intervention can be reduced and the efficiency of the use of nursing resources can be improved, thereby significantly improving the overall effect of neurology nursing. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments thereof made with reference to the following drawings:

[0049] Figure 1 It is a module schematic diagram of the analysis system based on the data model of the neurology nursing field of the present invention;

[0050] Figure 2 for Figure 1 Schematic diagram of the functional flow of the nursing behavior characteristics analysis module of neurology department. DETAILED DESCRIPTION

[0051] The technical system of the present invention is described clearly and completely below in conjunction with the accompanying drawings. 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 technicians in this field without creative work are within the scope of protection of the present invention.

[0052] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor systems and / or microcontroller systems.

[0053] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0054] To achieve this, please refer to Figure 1 to Figure 2 The present invention provides an analysis system based on a data model in the field of neurology nursing, the system comprising the following modules:

[0055] The neurology nursing behavior characteristic analysis module is used to obtain the neurology patient behavior dimension data and the neurology nursing record dimension data, and use the preset neurology nursing field data model to perform condition nursing behavior characteristic analysis on the neurology patient behavior dimension data and the neurology nursing record dimension data to obtain the neurology condition nursing behavior characteristics, including the behavior activity distribution loss and symptom onset probability of the neurology condition in each nursing record period;

[0056] The neurology nursing risk assessment module is used to calculate the neurology nursing risk assessment based on the behavioral activity distribution loss and symptom onset probability of the neurology condition in each nursing record period to obtain the neurology condition nursing risk assessment value;

[0057] The nursing demand and characteristic change law analysis module is used to obtain the nursing demand of neurology patients, and based on the nursing demand of neurology patients, conduct characteristic change law mining and analysis on the behavior activity distribution loss and symptom onset probability of neurology conditions in each nursing record period, so as to obtain the nursing change law between neurology nursing demand and behavior activity distribution loss and symptom onset probability;

[0058] The neurology nursing decision support module is used to perform nursing decision support analysis on the corresponding neurology patient nursing process based on the nursing change rules between neurology nursing needs and behavioral activity distribution loss and symptom onset probability, as well as the neurology disease nursing risk assessment value, and generate neurology patient nursing decision support recommendation plans to implement corresponding neurology patient nursing decision intervention measures.

[0059] In the embodiment of the present invention, please refer to Figure 1 FIG. 1 is a schematic diagram of a module of an analysis system based on a data model in the field of neurology nursing according to the present invention. In this example, the analysis system based on a data model in the field of neurology nursing includes the following modules:

[0060] S1: Neurology nursing behavior characteristic analysis module, used to obtain neurology patient behavior dimension data and neurology nursing record dimension data, and use the preset neurology nursing field data model to perform condition nursing behavior characteristic analysis on the neurology patient behavior dimension data and neurology nursing record dimension data, so as to obtain neurology condition nursing behavior characteristics, including the distribution loss of behavior activities and the probability of symptom onset of neurology condition in each nursing record period;

[0061] In an embodiment of the present invention, the purpose of designing and applying a neurology patient behavior self-assessment scale is to quantitatively evaluate the patient's behavior to obtain the patient's behavioral dimension data. The scale is designed based on common behavioral problems in the field of neurology, focusing on physiological behavioral changes (such as changes in motor skills, sleep patterns, eating conditions, etc.) and psychological behavioral changes (such as mood swings, cognitive abilities, anxiety and depression symptoms, etc.). During the design process, each behavioral indicator is quantified, and a clear scoring standard is defined. A Likert scale (such as 1 to 5 points, 1 for "never" to 5 for "always") is used to evaluate the patient's behavioral state, thereby obtaining the neurology patient's behavioral dimension data, including the neurology patient's physiological behavioral changes and the neurology patient's psychological behavioral changes. By extracting nursing record dimensional data from the electronic health record system (EHR) of the Department of Neurology, the EHR system contains detailed patient nursing records, including but not limited to nursing frequency, nursing content, patient response and other data. First, all relevant nursing records will be automatically screened out based on the patient's basic information and medical record number. The nursing frequency mainly reflects the number and cycle of nursing activities, such as how many times a day, the duration of each nursing, etc.; the nursing content includes specific nursing activities, such as drug management, body position changes, psychological counseling, etc. The patient response part records the patient's subjective experience and objective response during the nursing process, such as pain scores, emotional changes or adaptive feedback on nursing operations, thereby obtaining the dimensional data of neurology nursing records.At the same time, by performing time series analysis on the behavioral dimension data and nursing record dimension data of neurological patients obtained from the previous analysis, time series data of patient behavior and nursing records are generated. First, the collected patient behavior dimension data and nursing record dimension data are sorted according to timestamps. For each nursing event (such as the time when the nursing operation occurs), the corresponding patient behavior dimension data (such as the patient's psychological state, physiological condition, etc. at that time point) are associated. In the time series analysis, the stationarity test method (such as ADF test) is used to determine whether the data is stationary, and differential or smoothing processing is performed when necessary. Then, the data is modeled by the ARIMA (autoregressive integrated moving average) model or the LSTM (long short-term memory) model to capture the trend and periodicity of patient behavior and nursing records over time, and obtain detailed time series analysis results. A data model for neurological nursing is constructed by using a recurrent neural network (RNN) to perform dynamic distribution analysis. First, the neurological patient behavior dimension time series and nursing record dimension time series obtained in the previous steps are input into the RNN model for training. The model will automatically learn and identify the dependencies between the data based on the behavior and nursing record data at each moment input in the time series data. For different The nursing record time period is divided into multiple time windows. The data in each window will be processed by the RNN layer to output the dynamic distribution of patient behavior and nursing response in different time periods. Then, the nursing process of neurological patients in the corresponding nursing record period is analyzed based on the previously obtained dynamic distribution analysis results. First, according to the dynamic distribution of patient behavior and nursing response obtained in different nursing periods, the patient's behavioral activity distribution loss and symptom onset probability in a specific nursing period are analyzed. For example, the patient's anxiety and depression symptoms in a certain nursing period will be more obvious. The discomfort in the nursing response or the fluctuation of the patient's mood corresponds to a higher risk of illness. By combining clinical data and historical case data, the changes in the illness are quantitatively analyzed. Statistical methods such as regression analysis and chi-square test are used to analyze the correlation between nursing behavior and changes in the illness in different nursing periods, and the probability of symptom onset is calculated. In specific operations, the changes in the illness state are also associated with the timeliness of nursing activities by combining the time window of the nursing record, and the distribution loss results of the nursing behavior characteristics of the illness in different time periods are obtained. Finally, the behavioral activity distribution loss and symptom onset probability of the neurological illness in each nursing record period are obtained.

[0062] S2: Neurology nursing risk assessment module, used to calculate the neurology nursing risk assessment based on the behavioral activity distribution loss and symptom onset probability of the neurology condition in each nursing record period, and obtain the neurology condition nursing risk assessment value;

[0063] In an embodiment of the present invention, a suitable risk assessment calculation formula is formed by combining the quantity parameters of the nursing record period, the time variable parameters, the symptom onset probability, the symptom onset risk weight coefficient, the nursing behavior activity distribution intensity, the activity distribution intensity risk weight coefficient, the behavior activity distribution loss, the nursing behavior risk weight coefficient and related parameters to perform nursing risk assessment calculations, so as to quantify the risk measurement corresponding to the corresponding neurological disease nursing process, and finally obtain the neurological disease nursing risk assessment value.

[0064] S3: Nursing demand and characteristic change law analysis module, which is used to obtain the nursing demand of neurology patients, and based on the nursing demand of neurology patients, conduct characteristic change law mining and analysis on the behavior activity distribution loss and symptom onset probability of neurology conditions in each nursing record period, so as to obtain the nursing change law between neurology nursing demand and behavior activity distribution loss and symptom onset probability;

[0065] In an embodiment of the present invention, relevant nursing needs information of neurology patients is obtained through an integrated electronic health record system (EHR), which includes the patient's basic medical history, disease progression, medical equipment usage, drug treatment plan, and special needs in daily care, so as to obtain the corresponding nursing needs of neurology patients. At the same time, by combining the previously determined nursing needs, a time-series correlation statistical analysis is performed on the behavioral activity distribution loss and symptom onset probability of the neurological disease in each nursing record period, so as to use a time series-based machine learning algorithm (such as LSTM or GRU) to model the patient's condition, nursing needs, behavioral activity distribution loss and symptom onset probability. The data of each nursing record period will be sorted in chronological order and hierarchical analysis will be performed in combination with the patient's nursing needs. For example, if the nursing needs of a certain period are acute care, the symptom onset probability and behavioral activity distribution loss of this period will be higher; if it is a routine nursing need, the change range of behavioral activities and symptoms will be smaller. In this way, the changing trend of behavioral activities in each nursing period and the correlation with symptom onset can be identified, so as to identify the correlation between different nursing periods and disease fluctuations by using statistical methods (such as correlation analysis, covariance analysis) and data mining techniques (such as cluster analysis, association rule mining, etc.). Then, the previously acquired time-series correlation is analyzed by deep learning models (such as convolutional neural networks CNN, deep neural networks DNN, etc.). Specifically, based on the temporal correlation analysis results of the nursing needs of patients in the Department of Neurology, the distribution loss of behavioral activities, and the probability of symptom onset, the changing trends and patterns of behavioral activities in different time periods are extracted. For example, when the patient is in a certain pathological state (such as epileptic seizure period, recovery period, etc.), the distribution of their behavioral activities will change significantly, and this change is closely related to the probability of symptom onset. This process will use feature selection methods (such as principal component analysis PCA, factor analysis, etc.) to extract the core features that affect nursing needs and map them to the nursing change laws, so as to obtain the main nursing factors that affect the patient's symptom onset and behavioral activity distribution loss. Through these feature analyses, a set of prediction models can be generated, which can accurately predict the patient's behavioral activities and symptom changes in the future nursing period under different nursing needs, and further optimize the nursing plan. At the end of this step, nursing staff can use these change laws to accurately intervene in patients, avoid unnecessary nursing intervention and waste of resources, and finally obtain the nursing change laws between the nursing needs of neurology and the distribution loss of behavioral activities and the probability of symptom onset.

[0066] S4: Neurology nursing decision support module is used to perform nursing decision support analysis on the corresponding neurology patient nursing process based on the nursing change rules between neurology nursing needs and behavioral activity distribution loss and symptom onset probability, as well as the neurology disease nursing risk assessment value, and generate neurology patient nursing decision support recommendation plans to implement corresponding neurology patient nursing decision intervention measures.

[0067] In an embodiment of the present invention, a comparative judgment of the risk assessment value of the corresponding neurology patient nursing process is performed by combining the previously quantified neurology condition nursing risk assessment value. When the assessment value is lower than the preset nursing risk threshold, it indicates that the patient's condition is relatively stable, no obvious nursing risk events are found during the nursing process, and no additional emergency nursing measures are required. If the assessment value is greater than or equal to the set nursing risk threshold, it indicates that the patient's condition has a large nursing risk event, involving the occurrence of complications or worsening of the condition. The nursing staff should pay close attention and take necessary intervention measures. After identifying the neurology patients with nursing risk events, further nursing decision support analysis is performed. First, in combination with the patient's individual nursing needs, loss of behavioral activity ability, and probability of symptom onset, a data analysis and prediction model is used to comprehensively analyze the patient's nursing change rules. This analysis process mainly relies on a neurology patient behavior pattern recognition model based on big data, which is combined with the historical patient medical record data. , treatment response, dynamic changes in nursing needs and other information, predict the changes in nursing needs and fluctuations in the patient's condition in the future. In the specific implementation process, the model will analyze the relationship between the patient's loss of daily functional activities (such as motor skills, language skills, cognitive abilities, etc.) and the possibility of symptom recurrence. For example, for stroke patients, the model will calculate the risk of symptom aggravation in the short term in the future based on their current recovery status and past medical history, combined with the changing trend of their daily nursing needs, and output personalized nursing recommendations based on these predictions. On this basis, it will combine the results of nursing risk assessment to generate personalized nursing decision-making recommendation plans, which include but are not limited to: adjusting the frequency and intensity of nursing interventions, optimizing nursing plans (such as increasing the duration of rehabilitation treatment or symptom monitoring), recommending adjustments to drug use plans or auxiliary equipment (such as the use of pressure beds, monitoring instruments, etc.), and providing personalized nursing plans to ensure that patients receive the most appropriate nursing interventions, and ultimately implement corresponding nursing decision-making interventions for neurological patients.

[0068] Furthermore, the neurology nursing behavior characteristic analysis module includes the following functions:

[0069] By designing and applying a behavioral self-assessment scale for neurology patients and conducting a quantitative assessment of patient behavior using the behavioral self-assessment scale, we can obtain behavioral dimension data of neurology patients, including physiological behavioral changes and psychological behavioral changes of neurology patients.

[0070] By obtaining the corresponding neurology nursing record dimension data from the neurology electronic health record system, including neurology nursing frequency, neurology nursing content and neurology nursing patient response;

[0071] Conduct time series analysis on the neurology patient behavior dimension data and the neurology nursing record dimension data to generate the neurology patient behavior dimension time series and the neurology nursing record dimension time series;

[0072] A recurrent neural network was used to construct a data model for neurology nursing, and the data model was used to perform dynamic distribution analysis on the time series of neurology patient behavior and neurology nursing record dimensions under different nursing record time segments, to obtain the dynamic distribution of neurology patient behavior and the dynamic distribution of neurology nursing patient response under each nursing record period;

[0073] Based on the dynamic distribution of neurology patients' behaviors and the dynamic distribution of neurology patients' responses in each nursing record period, the nursing behavior characteristics of neurology patients' nursing process in the corresponding nursing record period were analyzed to obtain the behavioral activity distribution loss and symptom onset probability of neurology disease in each nursing record period.

[0074] As an embodiment of the present invention, refer to Figure 2 As shown, Figure 1 The function flow diagram of the neurology nursing behavior characteristic analysis module in this embodiment includes:

[0075] S11: By designing and applying a behavioral self-assessment scale for neurology patients and conducting a quantitative assessment of patient behavior using the behavioral self-assessment scale, we can obtain behavioral dimension data of neurology patients, including physiological behavioral changes and psychological behavioral changes of neurology patients;

[0076] In the embodiment of the present invention, the purpose of designing and applying the neurology patient behavior self-assessment scale is to quantitatively evaluate the patient's behavior to obtain the patient's behavioral dimension data. The scale is designed based on common behavioral problems in the field of neurology, focusing on physiological behavioral changes (such as changes in motor ability, sleep patterns, eating conditions, etc.) and psychological behavioral changes (such as mood swings, cognitive ability, anxiety and depression symptoms, etc.). During the design process, each behavioral indicator is quantified and a clear scoring standard is defined. The Likert scale (such as 1 to 5 points, 1 means "never" to 5 means "always") is used to evaluate the patient's Behavioral status: patients fill out the self-assessment scale regularly during the nursing process. Doctors or nurses record the data based on the patients' completion and summarize the data. For physiological behavioral changes, regular assessments can be made by observing the patients' athletic ability, food intake, etc.; for psychological behavioral changes, the self-assessment scale is combined with the patient's subjective reflection and standardized psychological test scales (such as the Hamilton Depression Rating Scale, Self-Rating Anxiety Scale, etc.) to ensure the comprehensiveness and accuracy of the behavioral dimension data. Finally, the behavioral dimension data of neurology patients are obtained, including physiological behavioral changes in neurology patients and psychological behavioral changes in neurology patients.

[0077] S12: Obtain the corresponding neurology nursing record dimension data from the neurology electronic health record system, including neurology nursing frequency, neurology nursing content, and neurology nursing patient response;

[0078] In an embodiment of the present invention, by extracting nursing record dimensional data from the electronic health record system (EHR) of the neurology department, the EHR system contains detailed patient nursing records, including but not limited to nursing frequency, nursing content, patient response and other data. First, all relevant nursing records are automatically screened out according to the patient's basic information and medical record number. The nursing frequency mainly reflects the number and cycle of nursing activities, such as how many times a day, the duration of each nursing, etc.; the nursing content includes specific nursing activities, such as drug management, body position change, psychological counseling, etc. The patient response part records the patient's subjective experience and objective response during the nursing process, such as pain score, emotional changes or adaptive feedback to nursing operations. Through the system interface, the required data is extracted using methods such as SQL query, and structured processing is performed to ensure the integrity and timeliness of the data. These nursing record dimensional data will be used in subsequent steps to analyze the relationship between patient behavior dynamics and nursing response, and finally the neurology nursing record dimensional data is obtained, including neurology nursing frequency, neurology nursing content and neurology nursing patient response.

[0079] S13: Perform time series analysis on the neurology patient behavior dimension data and the neurology nursing record dimension data to generate the neurology patient behavior dimension time series and the neurology nursing record dimension time series;

[0080] In an embodiment of the present invention, time series analysis is performed on the behavioral dimension data and nursing record dimension data of neurological patients obtained by previous analysis to generate time series data of patient behavior and nursing records. First, the collected patient behavior dimension data and nursing record dimension data are sorted according to timestamps. For each nursing event (such as the time when the nursing operation occurs), the corresponding patient behavior dimension data (such as the patient's psychological state, physiological condition, etc. at that time point) is associated. In the time series analysis, a stationarity test method (such as an ADF test) is used to determine whether the data is stationary. If necessary, differential or smoothing processing is performed. Then, the data is modeled using an ARIMA (autoregressive integrated moving average) model or an LSTM (long short-term memory) model to capture the trend and periodicity of patient behavior and nursing records over time, and obtain a time series model. The patient's behavior dimension and the nursing record dimension can be dynamically compared in different time periods to form a detailed time series analysis result, and finally a neurological patient behavior dimension time series and a neurological nursing record dimension time series are generated.

[0081] S14: Use recurrent neural network to build a data model for neurology nursing, and use the data model for neurology nursing to perform dynamic distribution analysis on the time series of neurology patient behavior dimension and the time series of neurology nursing record dimension under different nursing record time segments, and obtain the dynamic distribution of neurology patient behavior and the dynamic distribution of neurology nursing patient response under each nursing record period;

[0082] In an embodiment of the present invention, a data model for neurology nursing is constructed by using a recurrent neural network (RNN) to perform dynamic distribution analysis. First, the time series of neurology patient behavior dimension and nursing record dimension obtained in the above steps are input into the RNN model for training. The model automatically learns and identifies the dependency relationship between the data based on the behavior and nursing record data at each moment input in the time series data. The data is divided into multiple time windows for different nursing record time periods. The data in each window is processed by the RNN layer to output the dynamic distribution of patient behavior and the dynamic distribution of nursing response in different time periods. In specific operations, LSTM or GRU (gated recurrent unit) is used to process data to prevent the gradient vanishing problem and improve the prediction accuracy of the time series. After multiple iterative training, the RNN model can perform high-precision prediction and analysis of the behavior dynamics and nursing response dynamics in each nursing time period, and obtain the changing rules of patient behavior and the changing trends of nursing responses in different nursing time periods, and finally obtain the dynamic distribution of neurology patient behavior and the dynamic distribution of neurology nursing patient response in each nursing record time period.

[0083] S15: Based on the dynamic distribution of neurology patients' behaviors and the dynamic distribution of neurology patients' responses in each nursing record period, the nursing process of neurology patients in the corresponding nursing record period is analyzed to obtain the distribution loss of behavioral activities and the probability of symptom onset of neurology conditions in each nursing record period.

[0084] In an embodiment of the present invention, the nursing behavior characteristics of the neurological patient nursing process in the corresponding nursing record period are analyzed based on the previously obtained dynamic distribution analysis results. First, according to the dynamic distribution of patient behavior and the dynamic distribution of nursing response obtained in different nursing periods, the patient's behavioral activity distribution loss and the probability of symptom onset in a specific nursing period are analyzed. For example, the patient's anxiety and depression symptoms in a certain nursing period will be more obvious. Discomfort or patient mood fluctuations in nursing responses correspond to a higher risk of illness. By combining clinical data and historical case data, the changes in the illness are quantitatively analyzed. Statistical methods such as regression analysis and chi-square test are used to analyze the correlation between nursing behaviors and changes in the illness in different nursing periods, and the probability of symptom onset is calculated. In specific operations, the changes in the illness state are associated with the timeliness of nursing activities in combination with the time window of the nursing record, and the distribution loss results of the nursing behavior characteristics of the illness in different time periods are obtained. Finally, the behavioral activity distribution loss and the probability of symptom onset of the neurological illness in each nursing record period are obtained.

[0085] Furthermore, the analysis of the nursing behavior characteristics of the neurology patients in the corresponding nursing record period based on the dynamic distribution of the neurology patients' behaviors and the dynamic distribution of the neurology nursing patients' responses in each nursing record period includes:

[0086] The behavior activity spatiotemporal distribution matrix of neurology patients' behavior dynamic distribution in each nursing record period is analyzed to generate the behavior activity distribution matrix of neurology patients corresponding to each nursing record period, including the time distribution and spatial distribution of the patient's physiological and psychological behavior activities;

[0087] In an embodiment of the present invention, behavioral data of neurological patients in different nursing record time periods are collected, including data distribution corresponding to the patient's physiological behavior (such as body temperature, heart rate, respiratory rate, etc.) and psychological behavior (such as emotional state, cognitive ability performance, etc.), and the data of each nursing record time period are sorted and converted into a time and space distribution matrix of behavioral activities. At this time, the data can be obtained through monitoring equipment, questionnaires, medical records and other means. For the time distribution, the specific time points of the patient's behavioral activities in each time period are recorded, and the data are marked with corresponding time tags; the spatial distribution is calibrated according to the different environmental areas where the patient is located, such as wards, nursing stations, corridors and other areas. The generated spatiotemporal distribution matrix of behavioral activities will contain the distribution information of time and space under each nursing record time period, and finally the corresponding neurological patient behavior activity distribution matrix under each nursing record time period is generated.

[0088] Preferably, Fourier transform is used to perform behavioral activity fluctuation transformation analysis on the behavioral activity distribution matrix of neurology patients corresponding to each nursing record period, so as to obtain the behavioral activity change waveform of neurology patients corresponding to each nursing record period;

[0089] In an embodiment of the present invention, the previously obtained behavioral activity distribution matrix of each nursing record period is input into a Fourier transform model. Fourier transform is a mathematical tool used to convert time domain data into frequency domain data, which can help identify periodic fluctuations in the data. For the behavioral activity distribution matrix of each nursing record period, Fourier transform is performed on its physiological and psychological behaviors respectively. Through Fourier transform, the time series data is converted from the time domain to the frequency domain, and the changes in different frequency components are analyzed to identify the fluctuation pattern of the patient's behavioral activities. For example, changes in certain physiological behaviors show specific frequency fluctuations, while changes in psychological behaviors show lower frequency fluctuations. The data after Fourier transform will provide frequency domain information for subsequent fluctuation feature analysis, revealing the trend of behavioral activity changes of patients in different nursing periods, and finally obtaining the corresponding behavioral activity change waveforms of neurological patients in each nursing record period.

[0090] Preferably, the fluctuation characteristics of the behavior activity change waveforms of the neurology patients corresponding to each nursing record period are analyzed to obtain the fluctuation characteristics of the behavior activities of the neurology patients corresponding to each nursing record period; the fluctuation index is quantitatively calculated according to the fluctuation characteristics of the behavior activities of the neurology patients corresponding to each nursing record period to obtain the fluctuation index of the behavior activities of the neurology patients corresponding to each nursing record period;

[0091] In an embodiment of the present invention, by performing fluctuation feature analysis based on the previously obtained behavior activity fluctuation waveform, this analysis mainly extracts the amplitude, frequency, phase and other features of the waveform to understand the regularity and stability of the patient's behavior activity fluctuation, and by analyzing the waveform in different nursing record periods, the characteristics of high-frequency fluctuation, low-frequency fluctuation and irregular fluctuation are identified, thereby obtaining the corresponding neurological patient behavior activity fluctuation features, specifically including the distribution amplitude of neurological patient behavior activity, the distribution frequency of neurological patient behavior activity and the distribution phase of neurological patient behavior activity. At the same time, these fluctuation features can be further quantified into a behavior activity fluctuation index to describe the stability and variability of the patient's behavior activity. For example, if the patient's behavior activity fluctuation frequency is high and the amplitude is large in a certain period of time, it indicates that the patient's condition is relatively unstable, otherwise it indicates that the patient's state is relatively stable. By quantitatively calculating these fluctuation features, the behavior activity fluctuation index is obtained, that is, e^-(distribution amplitude × sin (distribution frequency × time + distribution phase) + distribution amplitude × cos (distribution frequency × time + distribution phase)), and finally the corresponding neurological patient behavior activity fluctuation index under each nursing record period is obtained.

[0092] Preferably, the corresponding spatiotemporal distribution gradient of the behavior activity of the neurology patients is obtained through the behavior activity distribution matrix of the neurology patients in each nursing record period, and the behavior distribution loss evaluation and analysis of the nursing process of the neurology patients in the corresponding nursing record period is performed based on the corresponding neurology patient behavior activity fluctuation index and the spatiotemporal distribution gradient of the behavior activity of the neurology patients in each nursing record period, so as to obtain the behavior activity distribution loss of the neurology condition in each nursing record period;

[0093] In an embodiment of the present invention, the spatiotemporal distribution gradient of the behavioral activities in each nursing record period is calculated based on the behavioral activity distribution matrix of neurological patients obtained by previous analysis. This gradient reflects the degree of change of the patient's behavioral activities in different time periods. The calculation of the behavioral activity distribution gradient can adopt the difference method, that is, calculating the change between adjacent time periods, or obtaining a more refined spatiotemporal distribution through the interpolation method. Then, combined with the behavioral activity fluctuation index obtained by previous analysis, the behavioral activity fluctuations in different nursing record periods are evaluated. If the behavioral activity fluctuation index is high and the distribution gradient is large, it means that there is a large loss in the patient's nursing process and the nursing measures need to be further optimized. Through this evaluation method, the patient's behavioral activity distribution loss in different nursing record periods can be quantified, that is, activity distribution loss = distribution gradient × e^-behavioral activity fluctuation index, and finally the behavioral activity distribution loss of the neurological condition in each nursing record period is obtained.

[0094] Preferably, based on the dynamic distribution of neurology nursing patient responses in each nursing record period, a quantitative analysis of the probability of symptom onset of the neurology patient's condition in the corresponding nursing record period is performed to obtain the probability of symptom onset of the neurology condition in each nursing record period.

[0095] In an embodiment of the present invention, a quantitative analysis of the probability of symptom onset of a disease symptom is performed based on the dynamic distribution data of the reactions of neurological patients obtained through previous analysis. At this time, the dynamic reaction data of the patient includes the fluctuations of the patient's physiological and psychological states during different nursing record periods. The probability of symptom onset in different nursing record periods is calculated through statistical models (such as Bayesian networks, Markov processes, etc.). These symptoms include but are not limited to neurological symptoms such as epileptic seizures, impaired consciousness, and mood swings. By modeling the frequency and changing patterns of the patient's symptoms, the probability of onset of each type of symptom in each time period is calculated. For example, if the patient has high-frequency mood swings in a certain time period, and such fluctuations are related to known symptoms of neurological diseases, the probability of symptom onset in that time period can be inferred, and finally the probability of symptom onset of the neurological disease in each nursing record period is obtained.

[0096] Furthermore, the behavior distribution loss evaluation analysis of the nursing process of neurology patients in the corresponding nursing record period based on the behavioral activity fluctuation index of neurology patients in each nursing record period and the spatiotemporal distribution gradient of the behavioral activity of neurology patients includes:

[0097] Obtain the ideal distribution target of neurology patients' behavioral activities corresponding to each nursing record period;

[0098] In an embodiment of the present invention, by defining the "ideal distribution target of behavioral activities of neurological patients", that is, under ideal conditions, the behavioral activity patterns that patients should exhibit in different nursing record time periods, these behavioral activities include but are not limited to the patient's autonomous movement, participation in treatment activities, changes in psychological state, etc., in specific implementation, a data model is established using the historical nursing data of neurological patients and relevant clinical research literature to clarify the ideal distribution targets of each behavioral activity in different nursing record time periods. For example, the nursing record time period can be divided by hour or each nursing shift, and standardized analysis is performed on the patient's behavioral activities in each time period to obtain the expected range of patient behavioral activities in each time period. The acquisition of this ideal distribution target can rely on statistical methods (such as mean, standard deviation, etc.) to summarize a large amount of patient data, and further adjust based on characteristics such as age, disease type, and treatment method, so that the behavioral activity targets in each nursing time period have reasonable standard values, and finally the corresponding ideal distribution targets of neurological patient behavior activities in each nursing record time period are obtained.

[0099] Preferably, based on the spatiotemporal distribution gradient of the behavior activities of neurology patients corresponding to each nursing record period, the behavior activity difference analysis is performed on the corresponding ideal distribution target of the behavior activities of neurology patients, so as to obtain the degree of difference between the behavior activities of neurology patients corresponding to each nursing record period and the ideal distribution target;

[0100] In an embodiment of the present invention, by comparing the difference between the actual patient behavior activities and the ideal distribution target in each nursing record period, in specific operations, firstly, the behavior activity data of the patient in each nursing period is subjected to spatiotemporal distribution analysis, which can be achieved by using location tracking technology (such as sensors or wearable devices) to obtain the patient's specific behavior path, moving distance, activity frequency and other data in each nursing period, and then calculating its spatiotemporal distribution gradient (such as the spatial change speed of the behavior activity in the time period, the change in activity intensity, etc.), and then by comparing these spatiotemporal distribution gradients with the ideal distribution target, a difference analysis method (such as mean square error or relative error) is used to evaluate the degree of difference in the patient's behavior activities. This analysis not only helps to identify deviations in patient behavior, for example, in a certain nursing period, if the patient should undergo more rehabilitation training, but the actual activity frequency is low, then the difference analysis results will reveal the difference between this behavior activity and the ideal target, and finally the degree of difference between the behavior activities of the corresponding neurological patients and the ideal distribution target in each nursing record period is obtained.

[0101] Preferably, a behavioral distribution loss assessment and analysis of the nursing process of neurological patients in corresponding nursing record periods is performed based on the behavioral activity fluctuation index of neurological patients in each nursing record period and the degree of difference between the behavioral activity of neurological patients and the ideal distribution target, so as to obtain the behavioral activity distribution loss of neurological conditions in each nursing record period.

[0102] In an embodiment of the present invention, by calculating the patient's behavioral activity fluctuation index in each nursing record period, the fluctuation index can usually be determined by analyzing the degree of fluctuation in the intensity or frequency of the patient's behavioral activities. If the patient's behavioral activities fluctuate violently in a certain period of time, the fluctuation index is higher; if the behavioral activities are stable, the fluctuation index is lower. This fluctuation index reflects the stability and consistency of the patient's behavior, and by combining the fluctuation index with the previously obtained degree of difference in behavioral activities, a behavioral distribution loss assessment analysis is performed. Specifically, through weighted average method or multiple regression analysis, the fluctuation index and the degree of difference are combined to form a comprehensive scoring model to assess the patient's behavioral activity loss in each nursing period. The greater the loss, the greater the gap between the patient's behavioral activities in this period and the ideal goal, and the higher the degree of irregularity or abnormality in the nursing process. This assessment result can help nursing staff to promptly discover potential problems in the patient's nursing process, such as whether the nursing method is appropriate, whether the patient has behavioral activity deviations due to fluctuations in the condition, etc., and finally obtain the behavioral activity distribution loss of the neurological condition in each nursing record period.

[0103] Furthermore, the quantitative analysis of the probability of symptom onset of the neurology patient nursing process in the corresponding nursing record period based on the dynamic distribution of the neurology patient nursing response in each nursing record period includes:

[0104] The dynamic distribution of patient responses in neurology nursing at each nursing record period is analyzed to obtain the corresponding neurology patient response symptom evolution map at each nursing record period;

[0105] In an embodiment of the present invention, by collecting the response data of neurological patients in each nursing record period, these data usually come from medical records, nursing logs and real-time monitoring data, including but not limited to the patient's physiological parameters (such as heart rate, blood pressure, respiratory rate, etc.), symptoms (such as headache, dizziness, epileptic seizures, etc.), drug use and nursing intervention, etc., and using statistical analysis methods or machine learning algorithms to process the data of each nursing period, and identify the regular changes of patient responses. Specifically, using time series data analysis methods, such as dynamic time warping (Dynamic TimeWarping, DTW) or long short-term memory network (LSTM), the patient response of each nursing record period is modeled, so as to depict the symptom evolution trend and dynamic distribution of neurological patients in different nursing periods. Based on the analysis results, a symptom evolution change map of neurological patients in each nursing period is generated. These maps show the trend of patient response changes over time during the nursing process, and finally the corresponding neurological patient response symptom evolution change map in each nursing record period is obtained.

[0106] Preferably, the fluctuation amplitude of the reaction symptoms of neurology patients corresponding to each nursing record period is analyzed to obtain the fluctuation amplitude of the reaction symptoms of neurology patients corresponding to each nursing record period;

[0107] In an embodiment of the present invention, based on the previously obtained patient response symptom evolution change map, a fluctuation analysis technique is used to calculate the fluctuation amplitude of the patient's symptoms in each nursing period. This process can be quantified through a mathematical model. Common methods include Fourier transform, maximum fluctuation amplitude calculation, root mean square fluctuation (RMS), etc. Fourier transform can convert time series signals into frequency domain signals to help identify periodic fluctuations in patient symptoms, while the maximum fluctuation amplitude and root mean square fluctuation can be used to evaluate the overall amplitude of patient symptom changes and their stability. During implementation, a custom algorithm is used to analyze the symptom data in each nursing record period, calculate the maximum and minimum fluctuation values ​​of the symptoms in the period, and further obtain the standard deviation of the fluctuation amplitude or other related statistics. In this way, the fluctuation amplitude of the patient's response in each nursing period can be quantified, and finally the corresponding fluctuation amplitude of the neurological patient's response symptoms in each nursing record period can be obtained.

[0108] Preferably, based on the fluctuation amplitude of the symptom reaction of the neurology patients in each nursing record period, the corresponding symptom reaction evolution change map of the neurology patients is analyzed for symptom outbreak nodes, so as to obtain the corresponding symptom outbreak point set of the neurology patients in each nursing record period;

[0109] In an embodiment of the present invention, the outbreak nodes of symptoms in each nursing record period are further analyzed based on the fluctuation amplitude of the symptoms of neurological patients previously obtained. The symptom outbreak points usually refer to the time nodes when the patient's symptoms change dramatically. These nodes are associated with specific disease attacks or nursing interventions. In order to identify these outbreak nodes, a mutation detection algorithm or an inflection point detection algorithm (such as change point detection, CUSUM algorithm, etc.) is used to perform a detailed analysis of the evolution change map of the patient's symptoms. These algorithms can identify points of dramatic changes in time series data and mark the time nodes when the symptoms rapidly change from a stable state to an acute or aggravated state. In specific operations, data mining technology is used to perform volatility analysis on the data of each nursing period to find out the sharp change points of symptom reactions and form a symptom outbreak point set. The resulting outbreak point set will help identify the key disease onset moments of patients in the nursing process, and finally obtain the corresponding neurological patient symptom outbreak point set under each nursing record period.

[0110] Preferably, based on the corresponding neurological patient response symptom outbreak point set in each nursing record period, the corresponding neurological patient response symptom evolution change map is subjected to a quantitative analysis of the probability of symptom onset of the disease, so as to obtain the probability of symptom onset of the neurological disease in each nursing record period.

[0111] In an embodiment of the present invention, by utilizing a previously identified symptom outbreak point set, the patient's response symptom evolution change map of each nursing period is combined with the patient's actual clinical condition, so that the probability of a patient experiencing a symptom outbreak in a certain period of time in each nursing period can be calculated through statistical analysis methods (such as Bayesian inference, Markov process, etc.). The specific operation includes dividing the symptom outbreak point set into time periods, constructing a probability model based on the patient's symptom fluctuation amplitude and related clinical data, and quantifying the possibility of symptom onset in each time period. To achieve this goal, Monte Carlo simulation and other methods can be used for multiple random sampling to evaluate the risk of symptom onset in patients. In the quantitative analysis process, it is also necessary to adjust the calculation model of the probability of onset in combination with factors such as individual differences, previous medical history, and drug use of patients. Through these analyses, the probability of symptom onset of neurological patients in each nursing record period is obtained, and finally the probability of symptom onset of neurological conditions in each nursing record period is obtained.

[0112] Further, the symptom outbreak node analysis of the corresponding neurology patient reaction symptom evolution change map based on the corresponding neurology patient reaction symptom fluctuation amplitude in each nursing record period includes:

[0113] The corresponding fluctuation rate of symptom reaction of neurology patients and the peak value of symptom reaction of neurology patients are obtained through the fluctuation amplitude of symptom reaction of neurology patients in each nursing record period;

[0114] In an embodiment of the present invention, a time series analysis is performed on the nursing records of neurology patients. The nursing record periods are divided according to the recorded time, usually in units of hours, days or weeks. The symptoms (such as headache, dizziness, limb numbness, etc.) of the neurology patients in each nursing period are quantitatively evaluated, and the intensity changes of these symptoms are recorded. The quantification method involves using a 0-10 point scoring system, or scoring according to a medical standardized scale, and calculating the amplitude of symptom fluctuations in each nursing record period, that is, the difference between the maximum and minimum values ​​of the symptom score. Then, the fluctuation amplitude data of each period is used to calculate the fluctuation rate using the fluctuation rate calculation formula (fluctuation rate = (standard deviation / mean value)×100%) to obtain the fluctuation rate in each period. The fluctuation rate reflects the degree of symptom fluctuation. The larger the value, the more severe the symptom fluctuation. In addition, in order to analyze the peak value of symptom fluctuation, the kurtosis analysis method is used (usually to calculate the high peak characteristic of the data). Specifically, by calculating the kurtosis value of the symptom score in each nursing period, the steepness of its distribution is evaluated. The higher the kurtosis value, the higher the peak of the symptom fluctuation, suggesting the existence of abnormal fluctuations or explosive symptom changes. This calculation usually uses the "kurtosis formula" in statistics. Finally, we can obtain the fluctuation change rate of symptoms in neurology patients and the peak value of symptoms in neurology patients.

[0115] Preferably, the corresponding resonance time point of the neurology patient's reaction symptoms is obtained through the corresponding neurology patient's reaction symptom evolution change map in each nursing record period, and the fluctuation peak correlation analysis is performed on the corresponding neurology patient's reaction symptom fluctuation kurtosis value in each nursing record period based on the resonance time point of the neurology patient's reaction symptoms, so as to obtain the corresponding neurology patient's reaction symptom fluctuation peak correlation factor in each nursing record period;

[0116] In an embodiment of the present invention, by modeling the evolution change map of the reaction symptoms of neurological patients, the changing trend of symptoms at different time points is captured. The change map is usually constructed by using time series analysis methods, such as local weighted regression (LOWESS) or smoothing filtering, to construct a dynamic change curve of the symptom score. The curve represents the change process of symptoms over time, wherein the resonant time point refers to the time node where the symptom fluctuation has a significant turning point or periodic change. These time nodes indicate that the symptoms have a drastic change or are associated with other clinical events. For example, if the symptom fluctuation shows a trend of periodic increase or sudden increase within a certain period of time , and is associated with the patient's other health conditions, then this period can be considered a resonance time point. After determining the resonance time point, a correlation analysis of the fluctuation kurtosis values ​​is performed based on these time points to determine the symptom fluctuation kurtosis values ​​before and after each resonance time point, and a correlation analysis is performed between these kurtosis values ​​and the resonance time point. This analysis can be completed through the Pearson correlation coefficient or linear regression model, and aims to evaluate whether the drastic changes in symptom fluctuations have a certain statistically significant relationship with the resonance time point. Through this analysis, the degree of correlation of symptom fluctuations can be revealed, and finally the corresponding correlation factor of the symptom fluctuation peak of neurology patients under each nursing record period is obtained.

[0117] Preferably, based on the corresponding neurology patient reaction symptom fluctuation peak correlation factor in each nursing record period, the corresponding neurology patient reaction symptom fluctuation change rate and neurology patient reaction symptom fluctuation peak value are analyzed to obtain the neurology patient reaction symptom fluctuation outbreak threshold value in each nursing record period;

[0118] In an embodiment of the present invention, the corresponding neurological patient reaction symptom fluctuation change rate and neurological patient reaction symptom fluctuation kurtosis value are further analyzed by combining the fluctuation peak correlation factor obtained by previous evaluation. These factors reflect the relationship strength between the fluctuation kurtosis of neurological patient reaction symptoms and the resonance time point within a specific nursing record period. Therefore, it is necessary to combine these factors with the fluctuation change rate and the fluctuation kurtosis value, and judge whether the symptoms have reached the critical point of fluctuation outbreak through comprehensive analysis. The key to the fluctuation outbreak threshold analysis is to determine a threshold, which is usually determined based on statistical analysis, that is, symptom fluctuation change rate × e^ (-fluctuation peak correlation factor × symptom fluctuation kurtosis value), and use it as the outbreak threshold. For example, if the outbreak threshold of a nursing record period is higher than the 95% quantile, it can be determined that the fluctuation change in this period is an explosive fluctuation, and finally the corresponding neurological patient reaction symptom fluctuation outbreak threshold under each nursing record period is obtained.

[0119] Preferably, based on the outbreak threshold of the fluctuation of the neurological patients' reaction symptoms corresponding to each nursing record period, the reaction symptom change values ​​in the corresponding neurological patients' reaction symptom evolution change map are marked and judged for the symptom outbreak nodes to obtain the corresponding neurological patients' reaction symptom outbreak point set in each nursing record period.

[0120] In an embodiment of the present invention, the reaction symptom change value at each time node in the reaction symptom evolution change map of neurological patients is marked according to the fluctuation outbreak threshold value obtained by previous quantification. Specifically, based on the corresponding symptom fluctuation outbreak threshold, each change point in the symptom evolution map is judged. If the symptom fluctuation amplitude or change rate at a certain time point exceeds the threshold, then the time point is considered to be a symptom outbreak node. In order to realize this marking process, it is first necessary to check each time period in the evolution map one by one by comparing the threshold with the symptom fluctuation data. If the symptom change in a certain time period exceeds the threshold, an "outburst point" mark is added in the time period, and the specific time and symptom value of the node are recorded. In addition, data visualization tools can be used to graphically mark these outbreak points on the symptom change curve, and finally the corresponding neurological patient reaction symptom outbreak point set under each nursing record time period is obtained.

[0121] Furthermore, the neurology nursing risk assessment calculation in the neurology nursing risk assessment module is calculated by a nursing risk assessment calculation formula; wherein the nursing risk assessment calculation formula is specifically:

[0122] ;

[0123] In the formula, is the neurological condition nursing risk assessment value at time, is the total number of nursing record periods, is the item index of the nursing record period, For the The integral time range corresponding to each nursing record period, is the time variable parameter, For neurological diseases The probability of symptom onset during the nursing record period, is the risk weight coefficient for symptom onset, For neurological diseases The distribution intensity of nursing behavior activities in each nursing record period, is the risk weight coefficient of activity distribution intensity, For neurological diseases The distribution loss of behavioral activities under each nursing record period, is the nursing behavior risk weight coefficient, It is the correction factor for the nursing risk assessment value of neurological conditions.

[0124] The present invention obtains a nursing risk assessment calculation formula by using a specific mathematical model and after verification, which is used to perform nursing risk assessment calculations on the behavioral activity distribution loss and symptom onset probability of neurological disease in each nursing record period. The nursing risk assessment calculation formula can reflect the nursing risk of the patient's condition at different time points by quantifying multiple factors in different nursing record periods. This comprehensive assessment helps nursing staff better understand the patient's overall health status, and then guide nursing intervention. By evaluating the symptom onset probability, nursing activity intensity and behavioral activity distribution loss in different nursing periods, the potential risk of disease deterioration can be effectively predicted. The formula uses the symptom onset probability and nursing behavior activity intensity as the main input variables, combined with the corresponding risk weight coefficient, to comprehensively assess the patient's risk in each nursing record period. The symptom onset probability reflects the fluctuation of the patient's condition, while the nursing activity intensity reflects the degree of implementation of nursing measures. The combination of the two enables risk assessment to be flexibly adjusted under different nursing strategies, thereby providing a basis for formulating personalized nursing plans. Secondly, the behavioral activity distribution loss and nursing behavior risk weight coefficient in the formula can quantify the "loss" of behavioral activities in the patient's nursing process. The behavioral activity loss reflects the adverse effects of undesirable nursing behaviors (such as improper nursing, lack of necessary intervention, etc.) on the patient's condition during the nursing process. Combined with the risk weight coefficient, it can better assess the potential hazards of improper nursing behaviors and promote nursing staff to strengthen refined nursing in practice to avoid errors and omissions. In addition, by introducing the correction coefficient, the risk assessment model can be adaptively adjusted, which means that in the actual clinical environment, based on different disease characteristics, nursing scenarios or the professional quality of nursing staff, the nursing risk assessment can be flexibly adjusted to improve the accuracy and practicality of the assessment. The role of the correction coefficient is to correct some conventional assumptions in the formula to ensure that the assessment value can more accurately reflect the actual nursing needs and risks. The nursing risk assessment value changes with time and can dynamically reflect the risk of the disease and the nursing process. By real-time analysis of the nursing records and patient condition changes in each period, nursing staff can timely discover potential nursing risks, make warnings and adjust nursing measures in time. This dynamic monitoring and prediction function has high clinical value, especially when the neurological disease is volatile and uncertain. In summary, this formula fully considers the risk assessment value of neurological disease care at time , the total number of nursing record periods , item index of nursing record period , No. The integral time range corresponding to the nursing record period , time variable parameter , neurological disease is in the first The probability of symptom onset during the nursing record period , symptom onset risk weight coefficient , neurological disease is in the first The distribution intensity of nursing behavior activities in each nursing record period , risk weight coefficient of activity distribution intensity , neurological disease is in the first Loss of behavioral activity distribution in nursing record period , nursing behavior risk weight coefficient , correction factor of nursing risk assessment value of neurology disease , according to the time Nursing risk assessment value of neurological conditions The correlation between the above parameters constitutes a functional relationship This formula can realize the calculation process of neurology nursing risk assessment for the behavioral activity distribution loss and symptom onset probability of neurology conditions in each nursing record period. At the same time, the correction coefficient of the neurology nursing risk assessment value is The introduction of can be adjusted according to the errors that occur in the calculation process, thereby improving the accuracy and applicability of the nursing risk assessment calculation formula.

[0125] Furthermore, the nursing needs and characteristic change law analysis module includes the following functions:

[0126] Capture neurology patient care needs;

[0127] In an embodiment of the present invention, relevant nursing demand information of neurology patients is obtained through an integrated electronic health record system (EHR). This information includes the patient's basic medical history, disease progression, medical equipment usage, drug treatment plan, and special needs in daily care. In specific operations, the system will calibrate and classify the nursing needs of neurology patients, for example, dividing the patient's nursing needs into routine care, acute care, special care and other categories. Through artificial intelligence (AI) and natural language processing (NLP) technology, the patient's clinical data is extracted from the electronic medical record, and various assessment scales (such as neurological function assessment scale, quality of life assessment scale, etc.) are combined for data processing and demand analysis. The acquisition of nursing needs not only depends on structured data, but also needs to further improve the patient's nursing needs file through text mining of the patient's chief complaint, doctor's and nurse's nursing records. At the end of this step, a detailed nursing needs report can be generated for each patient, and finally the corresponding nursing needs of neurology patients are obtained.

[0128] Preferably, based on the nursing needs of neurology patients, a temporal correlation analysis is performed on the behavioral activity distribution loss and the symptom onset probability of the neurology condition in each nursing record period to obtain the temporal correlation relationship between the neurology nursing needs and the behavioral activity distribution loss and the symptom onset probability;

[0129] In an embodiment of the present invention, a time-series correlation statistical analysis is performed on the behavioral activity distribution loss and the symptom onset probability of the neurological condition in each nursing record period in combination with the previously determined nursing needs, so as to use a time series-based machine learning algorithm (such as LSTM or GRU) to model the patient's condition, nursing needs, behavioral activity distribution loss and symptom onset probability. The data of each nursing record period will be sorted in chronological order and hierarchical analysis will be performed in combination with the patient's nursing needs. For example, if the nursing needs of a certain period are acute care, the symptom onset probability and behavioral activity distribution loss of that period will be higher; if it is regular care, the symptom onset probability and behavioral activity distribution loss of that period will be higher. In this way, the changing trend of behavioral activities in each nursing period and the correlation with symptom onset can be identified, thereby providing data support for the formulation of nursing plans. Temporal association analysis uses statistical methods (such as correlation analysis, covariance analysis) and data mining techniques (such as cluster analysis, association rule mining, etc.) to identify the correlation between different nursing periods and disease fluctuations, and help infer which nursing activities will lead to distribution loss of behavioral activities or onset of symptoms, and finally obtain the temporal correlation between neurology nursing needs and distribution loss of behavioral activities and the probability of symptom onset.

[0130] Preferably, based on the temporal correlation between neurological nursing needs and behavioral activity distribution losses and symptom onset probabilities, the characteristic change patterns of behavioral activity distribution losses and symptom onset probabilities of neurological conditions in each nursing record period are mined and analyzed to obtain the nursing change patterns between neurological nursing needs and behavioral activity distribution losses and symptom onset probabilities.

[0131] In an embodiment of the present invention, a deep learning model (such as a convolutional neural network CNN, a deep neural network DNN, etc.) is used to extract features and mine change rules of previously acquired temporal correlation relationships. Specifically, based on the temporal correlation analysis results of the nursing needs of neurology patients and the loss of distribution of behavioral activities and the probability of symptom onset, the change trends and patterns of behavioral activities in different time periods are extracted. For example, when a patient is in a certain pathological state (such as an epileptic seizure period, a recovery period, etc.), the distribution of his behavioral activities will change significantly, and this change is closely related to the probability of symptom onset. This process will use feature selection methods (such as principal component analysis PCA, factor analysis, etc.) to extract the core features that affect nursing needs, and map them to the nursing change rules, so as to obtain the main nursing factors that affect the patient's symptom onset and loss of distribution of behavioral activities. Through these feature analyses, a set of prediction models can be generated, which can accurately predict the changes in the patient's behavioral activities and symptoms in the future nursing period under different nursing needs, and further optimize the nursing plan. At the end of this step, the nursing staff can accurately intervene in the patient through these change rules to avoid unnecessary nursing intervention and waste of resources, and finally obtain the nursing change rules between the nursing needs of neurology and the loss of distribution of behavioral activities and the probability of symptom onset.

[0132] Furthermore, the neurology nursing decision support module includes the following functions:

[0133] Based on the nursing risk assessment value of the neurology condition, a nursing risk comparison and judgment is made on the corresponding neurology patient nursing process. If the nursing risk assessment value of the neurology condition is less than the preset nursing risk threshold, it is considered that there is no nursing risk event in the corresponding neurology patient nursing process; if the nursing risk assessment value of the neurology condition is greater than or equal to the preset nursing risk threshold, it is considered that there is a nursing risk event in the corresponding neurology patient nursing process;

[0134] In an embodiment of the present invention, a risk assessment value of the corresponding neurology patient nursing process is compared and judged by combining the previously quantified neurology condition nursing risk assessment value. When the assessment value is lower than the preset nursing risk threshold, it indicates that the patient's condition is relatively stable, no obvious nursing risk events are found during the nursing process, and no additional emergency nursing measures are required. If the assessment value is greater than or equal to the set nursing risk threshold, it means that the patient's condition has a large nursing risk event, involving the occurrence of complications or worsening of the condition. The nursing staff should pay close attention and take necessary intervention measures. In specific implementation, the patient's data is entered in real time through the medical information management system, and the condition assessment result is automatically generated through the algorithm. It is compared with the set nursing risk threshold, so as to provide real-time feedback to the nursing staff on whether it is necessary to adjust the patient's nursing process or increase monitoring efforts.

[0135] Preferably, a nursing decision support analysis is performed on the nursing process of neurological patients with corresponding nursing risk events based on the nursing change rules between neurological nursing needs and behavioral activity distribution loss and symptom onset probability, and a nursing decision support recommendation plan for neurological patients is generated to implement corresponding nursing decision intervention measures for neurological patients, including comprehensively considering the nursing needs, disease risks and behavioral activity ability loss of neurological patients to recommend corresponding personalized nursing plans, optimize nursing plans and adjust nursing intervention measures.

[0136] In an embodiment of the present invention, after identifying neurological patients with nursing risk events, further nursing decision support analysis is performed. First, in combination with the patient's individual nursing needs, loss of behavioral activity ability, and the probability of symptom onset, a data analysis and prediction model is used to comprehensively analyze the patient's nursing change patterns. This analysis process mainly relies on a neurological patient behavior pattern recognition model based on big data. The model will combine historical patient medical records, treatment responses, dynamic changes in nursing needs and other information to predict changes in nursing needs and fluctuations in the patient's condition in the future. In the specific implementation process, the model will analyze the relationship between the patient's daily functional activity ability loss (such as motor ability, language ability, cognitive ability, etc.) and the possibility of symptom recurrence. For example, for stroke patients, the model will predict the patient's current recovery status and the possibility of symptom recurrence based on their current recovery status. Based on the patient's past medical history and the changing trend of their daily care needs, we estimate the risk of worsening symptoms in the short term in the future, and output personalized nursing recommendations based on these predictions. On this basis, we will combine the results of nursing risk assessment to generate personalized nursing decision-making recommendations. The plan includes but is not limited to: adjusting the frequency and intensity of nursing interventions, optimizing nursing plans (such as increasing the duration of rehabilitation treatment or symptom monitoring), recommending adjustments to medication regimens or auxiliary equipment (such as using pressure beds, monitoring instruments, etc.), and providing personalized nursing plans. For example, how to optimize daily care arrangements based on the patient's mobility, or how to adjust intervention measures based on fluctuations in the patient's symptoms. Through this plan, nursing staff can accurately adjust nursing workflows and strategies to ensure that patients receive the most appropriate nursing interventions, and ultimately implement corresponding nursing decision interventions for neurological patients.

[0137] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0138] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. An analysis system based on a data model in the field of neurology nursing, characterized in that: Includes the following modules: The neurology nursing behavior characteristic analysis module is used to obtain the neurology patient behavior dimension data and the neurology nursing record dimension data, and use the preset neurology nursing field data model to perform condition nursing behavior characteristic analysis on the neurology patient behavior dimension data and the neurology nursing record dimension data to obtain the neurology condition nursing behavior characteristics, including the behavior activity distribution loss and symptom onset probability of the neurology condition in each nursing record period; The neurology nursing risk assessment module is used to calculate the neurology nursing risk assessment based on the behavioral activity distribution loss and symptom onset probability of the neurology condition in each nursing record period to obtain the neurology condition nursing risk assessment value; The nursing demand and characteristic change law analysis module is used to obtain the nursing demand of neurology patients, and based on the nursing demand of neurology patients, conduct characteristic change law mining and analysis on the behavior activity distribution loss and symptom onset probability of neurology conditions in each nursing record period, so as to obtain the nursing change law between neurology nursing demand and behavior activity distribution loss and symptom onset probability; The neurology nursing decision support module is used to perform nursing decision support analysis on the corresponding neurology patient nursing process based on the nursing change rules between neurology nursing needs and behavioral activity distribution loss and symptom onset probability, as well as the neurology disease nursing risk assessment value, and generate neurology patient nursing decision support recommendation plans to implement corresponding neurology patient nursing decision intervention measures.

2. The analysis system based on the neurology nursing field data model according to claim 1 is characterized in that: The neurology nursing behavior characteristic analysis module includes the following functions: By designing and applying a behavioral self-assessment scale for neurology patients and conducting a quantitative assessment of patient behavior using the behavioral self-assessment scale, we can obtain behavioral dimension data of neurology patients, including data on physiological behavioral changes and psychological behavioral changes of neurology patients. By obtaining the corresponding neurology nursing record dimension data from the neurology electronic health record system, including neurology nursing frequency data, neurology nursing content data and neurology nursing patient response data; Conduct time series analysis on the neurology patient behavior dimension data and the neurology nursing record dimension data to generate the neurology patient behavior dimension time series and the neurology nursing record dimension time series; A recurrent neural network was used to construct a data model for neurology nursing, and the data model was used to perform dynamic distribution analysis on the time series of neurology patient behavior and neurology nursing record dimensions under different nursing record time segments, to obtain the dynamic distribution of neurology patient behavior and the dynamic distribution of neurology nursing patient response under each nursing record period; Based on the dynamic distribution of neurology patients' behaviors and the dynamic distribution of neurology patients' responses in each nursing record period, the nursing behavior characteristics of neurology patients' nursing process in the corresponding nursing record period were analyzed to obtain the behavioral activity distribution loss and symptom onset probability of neurology disease in each nursing record period.

3. The analysis system based on the neurology nursing field data model according to claim 2 is characterized in that: The analysis of the nursing behavior characteristics of the neurology patients in the corresponding nursing record period based on the dynamic distribution of the neurology patients' behaviors and the dynamic distribution of the neurology nursing patients' responses in each nursing record period includes: The behavior activity spatiotemporal distribution matrix of neurology patients' behavior dynamic distribution in each nursing record period is analyzed to generate the behavior activity distribution matrix of neurology patients corresponding to each nursing record period, including the time distribution and spatial distribution of the patient's physiological and psychological behavior activities; Fourier transform was used to analyze the behavioral activity fluctuation transformation of the distribution matrix of neurology patients' behavioral activities in each nursing record period, and the corresponding waveform of neurology patients' behavioral activities in each nursing record period was obtained; The fluctuation characteristics of the waveforms of the behavior and activity changes of neurology patients corresponding to each nursing record period are analyzed to obtain the fluctuation characteristics of the behavior and activity of neurology patients corresponding to each nursing record period; the fluctuation index is quantitatively calculated based on the fluctuation characteristics of the behavior and activity of neurology patients corresponding to each nursing record period to obtain the fluctuation index of the behavior and activity of neurology patients corresponding to each nursing record period; The corresponding spatiotemporal distribution gradient of the behavior activities of neurology patients is obtained through the behavior activity distribution matrix of neurology patients in each nursing record period, and the behavior distribution loss evaluation and analysis of the nursing process of neurology patients in the corresponding nursing record period is performed based on the corresponding neurology patient behavior activity fluctuation index and the spatiotemporal distribution gradient of the behavior activities of neurology patients in each nursing record period, so as to obtain the behavior activity distribution loss of neurology disease in each nursing record period; Based on the dynamic distribution of patient responses in neurology nursing in each nursing record period, a quantitative analysis of the probability of symptom onset of neurological disease in the nursing process of neurology patients in the corresponding nursing record period was performed to obtain the probability of symptom onset of neurological disease in each nursing record period.

4. The analysis system based on the neurology nursing field data model according to claim 3 is characterized in that: The fluctuation characteristics of the behavioral activities of neurology patients specifically include the distribution amplitude of the behavioral activities of neurology patients, the distribution frequency of the behavioral activities of neurology patients, and the distribution phase of the behavioral activities of neurology patients.

5. The analysis system based on the neurology nursing field data model according to claim 3 is characterized in that: The behavior distribution loss evaluation and analysis of the nursing process of neurology patients in the corresponding nursing record period based on the corresponding neurology patient behavior activity fluctuation index and the spatiotemporal distribution gradient of the neurology patient behavior activity in each nursing record period includes: Obtain the ideal distribution target of neurology patients' behavioral activities corresponding to each nursing record period; Based on the temporal and spatial distribution gradient of the behavior activities of neurology patients corresponding to each nursing record period, the behavior activity difference analysis of the corresponding ideal distribution target of the behavior activities of neurology patients was performed to obtain the degree of difference between the behavior activities of neurology patients corresponding to each nursing record period and the ideal distribution target; Based on the behavioral activity fluctuation index of neurological patients corresponding to each nursing record period and the degree of difference between the behavioral activities of neurological patients and the ideal distribution target, a behavioral distribution loss evaluation and analysis of the nursing process of neurological patients in the corresponding nursing record period was performed to obtain the behavioral activity distribution loss of neurological conditions in each nursing record period.

6. The analysis system based on the neurology nursing field data model according to claim 3 is characterized in that: The method of quantitatively analyzing the probability of symptom onset of the neurology patient nursing process in the corresponding nursing record period based on the dynamic distribution of the neurology patient nursing response in each nursing record period includes: The dynamic distribution of patient responses in neurology nursing at each nursing record period is analyzed to obtain the corresponding neurology patient response symptom evolution map at each nursing record period; The fluctuation range of the symptom evolution of neurology patients corresponding to each nursing record period is analyzed to obtain the fluctuation range of the symptom reaction of neurology patients corresponding to each nursing record period; Based on the fluctuation amplitude of the corresponding neurology patients' reaction symptoms in each nursing record period, the corresponding neurology patients' reaction symptom evolution change map is analyzed for symptom outbreak nodes, and the corresponding neurology patients' reaction symptom outbreak point set in each nursing record period is obtained; Based on the corresponding neurological patients' symptom outbreak point set in each nursing record period, the corresponding neurological patients' symptom evolution change map is used to quantitatively analyze the probability of symptom onset of the disease, so as to obtain the symptom onset probability of neurological disease in each nursing record period.

7. The analysis system based on the neurology nursing field data model according to claim 6 is characterized in that: The symptom outbreak node analysis of the corresponding neurology patient reaction symptom evolution change map based on the corresponding neurology patient reaction symptom fluctuation amplitude in each nursing record period includes: The corresponding fluctuation rate of neurology patients' reaction symptoms and the peak value of neurology patients' reaction symptoms fluctuation are obtained through the fluctuation amplitude of neurology patients' reaction symptoms in each nursing record period; The corresponding resonance time points of the neurology patients' reaction symptoms are obtained through the evolution change maps of the neurology patients' reaction symptoms in each nursing record period, and the fluctuation peak correlation analysis of the corresponding neurology patients' reaction symptom fluctuation kurtosis values ​​in each nursing record period is performed based on the resonance time points of the neurology patients' reaction symptoms, so as to obtain the corresponding neurology patients' reaction symptom fluctuation peak correlation factors in each nursing record period; Based on the corresponding neurology patients' reaction symptom fluctuation peak correlation factors in each nursing record period, the corresponding neurology patients' reaction symptom fluctuation change rate and neurology patients' reaction symptom fluctuation peak value are analyzed to obtain the corresponding neurology patients' reaction symptom fluctuation outbreak threshold in each nursing record period; Based on the corresponding outbreak threshold of neurological patients' reaction symptoms fluctuation in each nursing record period, the reaction symptom change value in the corresponding neurological patients' reaction symptom evolution change map is marked and judged for symptom outbreak nodes to obtain the corresponding neurological patients' reaction symptom outbreak point set in each nursing record period.

8. The analysis system based on the neurology nursing field data model according to claim 1 is characterized in that: The nursing needs and characteristic change law analysis module includes the following functions: Capture neurology patient care needs; Based on the nursing needs of neurology patients, a temporal correlation analysis was conducted on the behavioral activity distribution loss and symptom onset probability of neurology conditions in each nursing record period, and the temporal correlation between neurology nursing needs and behavioral activity distribution loss and symptom onset probability was obtained; Based on the temporal correlation between neurology nursing needs, behavioral activity distribution loss, and symptom onset probability, the characteristic change rules of behavioral activity distribution loss and symptom onset probability of neurology conditions in each nursing record period were mined and analyzed to obtain the nursing change rules between neurology nursing needs, behavioral activity distribution loss, and symptom onset probability.

9. The analysis system based on the neurology nursing field data model according to claim 1 is characterized in that: The neurology nursing decision support module includes the following functions: Based on the nursing risk assessment value of the neurology condition, a nursing risk comparison is made for the corresponding neurology patient nursing process. If the nursing risk assessment value of the neurology condition is less than the preset nursing risk threshold, it is considered that there is no nursing risk event in the corresponding neurology patient nursing process. If the nursing risk assessment value of the neurology disease is greater than or equal to the preset nursing risk threshold, it is considered that there is a nursing risk event in the nursing process of the corresponding neurology patient; Based on the nursing change rules between neurology nursing needs and behavioral activity distribution loss and symptom onset probability, a nursing decision support analysis is conducted on the nursing process of neurology patients with corresponding nursing risk events, and a nursing decision support recommendation plan for neurology patients is generated to implement corresponding nursing decision intervention measures for neurology patients, including comprehensively considering the nursing needs, disease risks and behavioral activity ability loss of neurology patients to recommend corresponding personalized nursing plans, optimize nursing plans and adjust nursing intervention measures.

Images