Neurology clinical nursing potential safety hazard analysis method and device

Abstract

The invention discloses a neurology department clinical nursing potential safety hazard analysis method and device, and relates to the technical field of clinical nursing potential safety hazard analys.The method comprises the steps that physiological data and behavior data of a patient are collected, and multi-source data are preprocessed through weighted Kalman filtering; performing multi-dimensional fusion on the processed data, calculating a comprehensive health score by adopting a weighted average method, and evaluating the potential safety hazard of the patient in combination with a convolutional neural network; and based on the comprehensive health score, generating a personalized nursing scheme, and providing nursing decision support for medical staff. According to the clinical nursing potential safety hazard analysis method for the neurology department, the physiological data and the behavior data are combined, multi-dimensional data fusion is carried out through weighted Kalman filtering and the convolutional neural network, the accuracy of patient health assessment is improved, noise and errors are reduced, and the consistency and reliability of the data are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of clinical nursing safety hazard analysis, and in particular to a method and device for analyzing safety hazard in clinical nursing of neurology. Background Art

[0002] In recent years, with the development of medical technology, clinical nursing work in neurology has gradually developed in the direction of digitalization and intelligence. The introduction of intelligent nursing solutions provides new ideas for improving nursing efficiency and safety. Through multi-dimensional data collection, analysis and processing, combined with advanced algorithm technology (such as Kalman filtering, convolutional neural network, etc.), medical staff can more accurately assess the health status of patients, identify potential safety hazards, and intervene in time. Data collection includes patients' physiological data (such as blood sugar, blood pressure, body temperature, heart rate, etc.) and behavioral data (such as motor ability, language ability, etc.), which provide basic support for personalized nursing in neurology. In recent years, many studies have begun to focus on how to improve the personalization and accuracy of nursing solutions through comprehensive data analysis, and intelligent technology has gradually penetrated into all aspects of nursing work.

[0003] However, although the existing neurology nursing methods can monitor and evaluate physiological data, the existing technologies still have some shortcomings in data analysis and processing, health risk assessment and personalized nursing plan generation. First, the existing systems lack effective data fusion methods when processing multi-source data. Many systems can only rely on a single data or indicator for evaluation, and cannot fully consider the patient's health status and its potential risks. Secondly, most traditional health assessment methods rely on manual input and simple threshold judgment, lacking dynamic real-time and intelligence. Although some systems have tried to introduce machine learning algorithms for health assessment, many methods only focus on the processing of physiological data, ignoring the combination and analysis of multi-dimensional information such as patient behavior and environmental factors. In addition, the existing technologies are relatively limited in their capabilities in hazard assessment and nursing plan generation. Most systems cannot automatically generate personalized nursing plans or provide real-time nursing advice, resulting in medical staff's decision-making still relying on traditional manual experience and manual operations. Therefore, the existing technologies cannot meet the complex needs of clinical nursing in neurology, especially in multi-dimensional data analysis and safety hazard identification. There are obvious technical bottlenecks. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problems solved by the present invention are: the existing clinical nursing methods in neurology lack multi-dimensional integration of data collection and analysis, and cannot fully reflect the patient's health status; health assessment and hidden danger prediction rely too much on traditional threshold judgment, lack of intelligence and real-time dynamic adjustment capabilities, and how to generate personalized nursing plans in a timely manner.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: a method for analyzing safety hazards in clinical nursing in neurology, comprising collecting physiological data and behavioral data of patients, preprocessing multi-source data through weighted Kalman filtering, and dynamically adjusting weighting coefficients according to the noise characteristics of different data sources by adaptive weighted Kalman filtering; multi-dimensionally fusing the processed data, assigning weights to each item of data according to the degree of impact on the patient's health and calculating a comprehensive health score, analyzing the patient's time-series physiological data and behavioral data in combination with a convolutional neural network, and extracting long-term health change trends and potential hidden dangers; generating personalized nursing plans based on the comprehensive health scores, and providing nursing decision support to medical staff.

[0007] As a preferred solution of the method for analyzing safety hazards in clinical nursing of neurology of the present invention, wherein: the pathological data collection includes collecting physiological data and behavioral data of the patient; Physiological data include blood sugar level, blood pressure, body temperature, heart rate, respiratory rate, electrocardiogram; Behavioral data include motor skills and language skills.

[0008] As a preferred solution of the method for analyzing safety hazards in clinical nursing of neurology of the present invention, wherein: the preprocessing of multi-source data by weighted Kalman filtering includes removing fluctuations caused by sensor errors and environmental noise by prediction and correction; The physiological data collected from multiple sources are processed using weighted Kalman filtering to estimate the current state value based on the sensor data at the previous moment, and the state estimate is adjusted according to the error between the current observed data and the predicted value; In weighted Kalman filtering, the estimated value of each sensor is weighted using its weight coefficient, and the weight of the sensor is determined by the ratio of the square of the signal-to-noise ratio to the standard deviation of the sensor; The weighted Kalman gain is applied to correct the differences between the observed and predicted values ​​of multi-source data; The result of weighted Kalman filtering is the state estimate at each moment.

[0009] As a preferred embodiment of the method for analyzing safety hazards in clinical nursing of neurology described in the present invention, the multi-dimensional fusion of the processed data includes first standardizing the data into a unit-less magnitude, and then using a weighted average method, and assigning weights to each data according to its importance to the patient's health status, and the weights are obtained through clinical experience or expert evaluation.

[0010] As a preferred solution of the method for analyzing safety hazards in clinical nursing in neurology described in the present invention, the extraction of long-term health change trends and potential hidden dangers includes reading the user's historical health data, screening out data types with data abnormalities, and setting them as key indicators, performing cyclic detection within a preset period, and tracking the patient's key indicator data in real time. When the patient has abnormal key indicators, an alarm is automatically triggered to indicate possible health hazards.

[0011] As a preferred embodiment of the method for analyzing safety hazards in clinical nursing of neurology of the present invention, the extraction of long-term health change trends and potential hazard risks also includes calculating the comprehensive health score of the patient, which is based on the patient's physiological data, behavioral data and environmental data, and each data is weighted according to the degree of impact on health; Calculate the contribution of the patient's behavioral data to the hidden dangers. The behavioral data includes motor ability, language ability, etc., and calculate them weightedly according to the weight coefficient; The health score, contribution of behavioral data and impact of environmental data are weighted and summed to obtain a weighted comprehensive value; The weighted comprehensive value is transformed nonlinearly using an activation function to obtain the final hidden danger risk score, which represents the potential safety hazards of the patient in the current state; The weight coefficient is determined through clinical data or expert evaluation, and the hidden danger risk score reflects the assessment result of the patient's potential risk based on the comprehensive health status.

[0012] As a preferred embodiment of the method for analyzing safety hazards in clinical nursing of neurology of the present invention, the method of generating a personalized nursing plan includes determining a nursing decision score based on a health score and a hazard risk score, automatically generating a personalized nursing plan, and the nursing decision score It is the basis for generating personalized care plans based on the patient's health score and hidden danger risk scores .

[0013] A device using the above-mentioned method for analyzing safety hazards in clinical nursing in neurology includes a data acquisition module, an evaluation module, and a decision-making module. The data acquisition module is used to collect the patient's physiological data and behavioral data, and pre-process the multi-source data through weighted Kalman filtering. The adaptive weighted Kalman filtering dynamically adjusts the weighting coefficient according to the noise characteristics of different data sources. The evaluation module is used to perform multi-dimensional fusion of the processed data, assign weights to each data according to the degree of impact on the patient's health and calculate the comprehensive health score, and analyze the patient's time-series physiological data and behavioral data in combination with a convolutional neural network to extract long-term health change trends and potential hidden dangers. The decision-making module is used to generate personalized nursing plans based on the comprehensive health score and provide nursing decision support to medical staff.

[0014] A computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement a method for analyzing safety hazards in clinical nursing of neurology.

[0015] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a method for analyzing safety hazards in clinical nursing of neurology.

[0016] Beneficial effects of the present invention: The method for analyzing safety hazards in clinical nursing in neurology provided by the present invention combines physiological data and behavioral data, and uses weighted Kalman filtering and convolutional neural networks (CNN) for multi-dimensional data fusion, thereby improving the accuracy of patient health assessment, reducing noise and errors, and ensuring data consistency and reliability. The deep learning model can assess the patient's hidden danger risk in real time, and the system dynamically adjusts the nursing plan according to the patient's health score and hidden danger risk score to ensure timely intervention measures and minimize potential health hazards. The present invention achieves better results in terms of reliability, timeliness and data consistency. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0018] Figure 1 An overall flow chart of a method for analyzing safety hazards in clinical nursing in neurology provided in the first embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0020] Example 1, reference Figure 1 , which is an embodiment of the present invention, provides a method for analyzing safety hazards in clinical nursing of neurology, comprising: S1: Collect the patient's physiological data and behavioral data, pre-process the multi-source data through weighted Kalman filtering, and the adaptive weighted Kalman filter dynamically adjusts the weighting coefficient according to the noise characteristics of different data sources.

[0021] Furthermore, pathological data collection includes collecting physiological data and behavioral data of patients.

[0022] Physiological data include blood sugar level, blood pressure, body temperature, heart rate, respiratory rate, and electrocardiogram, which come from medical devices such as smart bracelets, blood sugar monitors, thermometers, and electrocardiogram monitors.

[0023] Behavioral data include motor skills, language skills, and these data are obtained through sensors (such as accelerometers, motion detectors) and other non-invasive devices.

[0024] It should be noted that data processing includes removing fluctuations caused by sensor errors and environmental noise through prediction and correction. Weighted Kalman filtering is used to process physiological data collected from multiple sources. The state value at the current moment is estimated based on the sensor data at the previous moment, and the state estimate is adjusted according to the error between the current observed data and the predicted value. In weighted Kalman filtering, the estimated value of each sensor is weighted using its weight coefficient, and the weight of the sensor is determined based on the ratio of the square of the signal-to-noise ratio to the standard deviation of the sensor. The difference between the observed value and the predicted value of multi-source data is corrected by applying the weighted Kalman gain. The result of weighted Kalman filtering is the state estimate at each moment, expressed as: ; in, Indicates time The estimated value of the state, Indicates The weighting coefficients of the sensors are represents the sensor data weight, Indicates The observation data of sensors, represents the Kalman gain, represents the measurement matrix, Indicates The sensors in The estimated value of the time, Indicates the total number of data items.

[0025] Sensors and devices may generate errors and noise due to environmental factors or their own precision limitations, resulting in data fluctuations. In the process of multi-dimensional data fusion, in order to ensure the accuracy of the data, it is necessary to first denoise the time series data. We use the Kalman filter algorithm to deal with the noise in the time series data.

[0026] S2: Perform multi-dimensional fusion on the processed data, assign weights to each data item according to its impact on the patient's health and calculate a comprehensive health score. Combined with a convolutional neural network to analyze the patient's time-series physiological and behavioral data, extract long-term health change trends and potential risks.

[0027] Furthermore, multi-dimensional data fusion involves first standardizing the data into unitless magnitudes, and then using a weighted average method, where each piece of data is given a weight based on its importance to the patient's health status. The weight is obtained through clinical experience or expert evaluation and is expressed as: ; in, represents the patient's comprehensive health score, Indicates Physiological data items, Indicates The weight of a data item.

[0028] In an alternative embodiment of the present invention, for example, for stroke patients, blood glucose level may be more important, so its weight may be higher, while body temperature may be relatively less important. The standardization of blood glucose level converts the data into a dimensionless standard score by comparing it with the normal value range of blood glucose. Similarly, other data such as blood pressure and body temperature also need to be processed using corresponding standardized methods to ensure that they can be compared and integrated under a unified standard.

[0029] It should be noted that extracting long-term health change trends and potential risks includes reading the user's historical health data, screening out data types with data anomalies, setting them as key indicators, performing cyclic detection within a preset period, and tracking the patient's key indicator data in real time. When the patient has abnormalities in key indicators, an alarm is automatically triggered to indicate possible health risks.

[0030] It should also be noted that extracting long-term health change trends and potential hidden danger risks also includes calculating the patient's comprehensive health score. The score is based on the patient's physiological data, behavioral data and environmental data, and each data is weighted according to the degree of impact on health. Calculate the contribution of the patient's behavioral data to hidden dangers. Behavioral data includes motor skills, language skills, etc., and weighted calculation according to the weight coefficient. The health score, behavioral data contribution and environmental data impact are weighted and summed to obtain a weighted comprehensive value. Use an activation function to perform a nonlinear transformation on the weighted comprehensive value to obtain the final hidden danger risk score, which represents the patient's potential safety hazards in the current state. The weight coefficient is determined by clinical data or expert evaluation. The hidden danger risk score reflects the assessment result of the comprehensive health status to the patient's potential risks, expressed as: ; in, Indicates the hidden danger risk score, reflecting the patient's health risks. Representing physiological data Contribution to health risks, Indicates the impact of environmental data on patients’ health risks, Represents the weight coefficient corresponding to each data dimension, Represents the activation function.

[0031] Specifically, in order to accurately identify potential safety hazards of patients in different states, a convolutional neural network (CNN) is combined. CNN performs well in processing complex data such as images and time series data, so it is suitable for extracting deep features from multi-dimensional data.

[0032] The input layer receives multi-dimensional fused data, including the patient's health score, behavioral data, environmental data, etc. Each data item is standardized to ensure the uniform scale of the data.

[0033] Convolutional layers are used to extract features from input data and identify important patterns related to patients’ health risks. For example, physiological data and behavioral data may be combined into specific health risk patterns, and convolution operations extract these patterns through local receptive fields.

[0034] The pooling layer reduces the dimension of the data by downsampling operations and retains important features. Through pooling, the computational complexity can be effectively reduced and the robustness of the features can be maintained.

[0035] The fully connected layer fuses the pooled features to obtain a comprehensive hidden danger identification feature vector, which will provide a basis for subsequent hidden danger risk scoring.

[0036] The neurons in the output layer represent the hidden danger risk score. An activation function (such as sigmoid) is used to convert the output into a value between 0 and 1, representing the hidden danger risk score.

[0037] S3: Generate personalized care plans based on comprehensive health scores and provide care decision support to medical staff.

[0038] Furthermore, personalized nursing plans are provided to medical staff, including determining nursing decision scores based on health scores and hidden danger risk scores, and automatically generating personalized nursing plans and nursing decision scores. It is the basis for generating personalized care plans based on the patient's health score and hidden danger risk scores , expressed as: ; in, and Represent the weight coefficients of health score and hidden danger risk score respectively.

[0039] When the nursing decision score exceeds the preset value, an alarm is issued to inform the nursing staff to test key indicators. If the key indicators are within the normal range, other indicators are tested. If all indicators are within the normal range, the ward temperature is tested to see if it is abnormal and the attending physician is informed.

[0040] It should be noted that, in one embodiment of the present invention, a patient P003 had a continuous blood sugar fluctuation problem during hospitalization. Historical health data showed that his blood sugar level fluctuated greatly in the past two years, and high blood sugar often occurred at different times. Based on the historical data, blood sugar was set as a key indicator by the system. During the hospitalization, the system collected blood sugar data of patient P003 in real time every hour through a blood sugar monitor.

[0041] When the blood sugar level of patient P003 suddenly increased from 6.5 mmol / L to 9.0 mmol / L, the system detected the abnormal fluctuation through the preset threshold and sent an alarm to the medical staff according to the preset alarm mechanism.

[0042] The alert included: patient ID, current blood sugar value, preset blood sugar threshold, and recommended next steps (increasing insulin dose). Medical staff quickly responded to the alert results, adjusted the patient's insulin treatment plan, and continued to monitor blood sugar levels to ensure that the patient's condition was effectively managed.

[0043] Example 2, an embodiment of the present invention, provides a method for analyzing safety hazards in clinical nursing of neurology. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through simulation experiments.

[0044] First of all, this experiment was verified for stroke patients. At the beginning of the experiment, the physiological data and behavioral data of stroke patients were collected. Data collection uses wearable devices (such as smart bracelets, blood glucose monitors) and medical equipment in the ward (such as electrocardiogram monitors, blood pressure monitors, thermometers, etc.). The patient's physiological data includes: blood glucose level (unit: mmol / L), blood pressure (unit: mmHg), electrocardiogram data, body temperature (unit: °C), respiratory rate (unit: times / minute), etc. Behavioral data include motor ability (unit: minutes) and language ability (unit: minutes). All data are collected in real time at a set sampling frequency (1 time per minute). Because different types of sensors may have errors and noise, especially in stroke patients, sudden changes in the condition may cause data fluctuations, so we use weighted Kalman filtering to predict and correct the data to ensure the accuracy of the data.

[0045] Table 1 Experimental data comparison table Patient ID Blood glucose level (mmol / L) Blood pressure (mmHg) Body temperature (°C) Heart rate (beats / minute) Respiratory rate (times / minute) Sports ability score (points) Comprehensive health score (points) Hidden danger risk score (points) Nursing decision score (points) P001 6.8 150 / 95 37.2 85 18 6 76 82 80 P002 5.5 130 / 80 36.5 78 16 8 88 74 83 P003 8 160 / 100 37.5 95 20 5 70 85 75 P004 7.2 140 / 90 36.8 88 17 7 78 79 78 P005 9.5 170 / 110 38 100 22 4 65 90 70 According to the data in the table, it is obvious that the present invention has advantages in health assessment and hidden danger prediction. For patients with higher blood sugar levels (such as P003 and P005), the hidden danger risk score is significantly higher, indicating that these patients face higher health risks. By using the multidimensional data fusion technology in the present invention, the hidden danger risk score can accurately reflect the potential health risks of patients.

[0046] For example, the hidden danger risk score of patient P003 is 85, which means that he may face a greater health risk, so the system automatically generates a targeted nursing plan for him and prompts medical staff to test key indicators. In addition, the calculation of the nursing decision score makes the nursing plan more personalized by considering the weights of the health score and the hidden danger risk score. For patients P001 and P004, although their health scores are high, due to the high hidden danger risk scores, the system still formulates preventive nursing plans for them to ensure timely intervention.

[0047] Example 3, an embodiment of the present invention, provides a neurology clinical nursing safety hazard analysis device, including a data acquisition module, an evaluation module, and a decision module.

[0048] Among them, the data acquisition module is used to collect the patient's physiological data and behavioral data, and pre-processes multi-source data through weighted Kalman filtering. The adaptive weighted Kalman filtering dynamically adjusts the weighting coefficient according to the noise characteristics of different data sources; the evaluation module is used to perform multi-dimensional integration of the processed data, assign weights to each data according to the degree of impact on the patient's health and calculate the comprehensive health score. The convolutional neural network is used to analyze the patient's time-series physiological data and behavioral data, and extract long-term health change trends and potential risks; the decision-making module is used to generate personalized care plans based on the comprehensive health score, and provide care decision support to medical staff.

[0049] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0050] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0051] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk case (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0052] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0053] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should be included in the scope of the claims of the present invention.

Claims

1. A method for analyzing safety hazards in clinical nursing of neurology, characterized in that: include: Collect the patient's physiological and behavioral data, and pre-process the multi-source data through weighted Kalman filtering. The adaptive weighted Kalman filter dynamically adjusts the weighting coefficient according to the noise characteristics of different data sources; The processed data is integrated in multiple dimensions, each data is weighted according to its impact on the patient's health and a comprehensive health score is calculated. The patient's time-series physiological data and behavioral data are analyzed using a convolutional neural network to extract long-term health change trends and potential risks. Based on the comprehensive health score, a personalized care plan is generated and care decision support is provided to medical staff.

2. The method for analyzing safety hazards in clinical nursing of neurology according to claim 1, characterized in that: The collecting of the patient's physiological data and behavioral data includes collecting the patient's physiological data and behavioral data in real time; Physiological data include blood sugar level, blood pressure, body temperature, heart rate, respiratory rate, electrocardiogram; Behavioral data include motor skills and language skills.

3. The method for analyzing safety hazards in clinical nursing of neurology according to claim 2, characterized in that: The preprocessing of multi-source data by weighted Kalman filtering includes removing fluctuations caused by sensor errors and environmental noise by prediction and correction; The physiological data collected from multiple sources are processed using weighted Kalman filtering to estimate the current state value based on the sensor data at the previous moment, and the state estimate is adjusted according to the error between the current observed data and the predicted value; In weighted Kalman filtering, the estimated value of each sensor is weighted using its weight coefficient, and the weight of the sensor is determined by the ratio of the square of the signal-to-noise ratio to the standard deviation of the sensor; The weighted Kalman gain is applied to correct the differences between the observed and predicted values ​​of multi-source data; The result of weighted Kalman filtering is the state estimate at each moment.

4. The method for analyzing safety hazards in clinical nursing of neurology according to claim 3, characterized in that: The multi-dimensional fusion of the processed data includes first standardizing the data into a unit-less magnitude, and then using a weighted average method to assign a weight to each data item according to its importance to the patient's health condition, and the weight is obtained through clinical experience or expert evaluation.

5. The method for analyzing safety hazards in clinical nursing of neurology according to claim 4, characterized in that: The extraction of long-term health change trends and potential hidden risks includes reading the user's historical health data, screening out data types with data anomalies, and setting them as key indicators, performing cyclic detection within a preset period, and tracking the patient's key indicator data in real time. When the patient has abnormal key indicators, an alarm is automatically triggered to indicate possible health risks.

6. The method for analyzing safety hazards in clinical nursing of neurology according to claim 5, characterized in that: The extracting of long-term health change trends and potential hidden risks also includes calculating the patient's comprehensive health score, which is based on the patient's physiological data, behavioral data and environmental data, and each data is weighted according to the degree of impact on health; Calculate the contribution of the patient's behavioral data to the hidden dangers. The behavioral data includes motor ability and language ability, and the data is weighted and calculated according to the weight coefficient; The health score, contribution of behavioral data and impact of environmental data are weighted and summed to obtain a weighted comprehensive value; The weighted comprehensive value is transformed nonlinearly using an activation function to obtain the final hidden danger risk score, which represents the potential safety hazards of the patient in the current state; The weight coefficient is determined through clinical data or expert evaluation, and the hidden danger risk score reflects the assessment result of the patient's potential risk based on the comprehensive health status.

7. The method for analyzing safety hazards in clinical nursing of neurology according to claim 6, characterized in that: The generating of personalized nursing plan includes determining nursing decision score based on health score and hidden danger risk score, automatically generating personalized nursing plan, nursing decision score It is the basis for generating personalized care plans based on the patient's health score and hidden danger risk scores .

8. A device using the method for analyzing safety hazards in clinical nursing of neurology as claimed in any one of claims 1 to 7, characterized in that: Including data collection module, evaluation module and decision-making module; The data acquisition module is used to collect the patient's physiological data and behavioral data, and pre-process the multi-source data through weighted Kalman filtering. The adaptive weighted Kalman filtering dynamically adjusts the weighting coefficient according to the noise characteristics of different data sources; The evaluation module is used to perform multi-dimensional fusion of the processed data, assign weights to each data item according to its impact on the patient's health and calculate a comprehensive health score, and analyze the patient's time-series physiological data and behavioral data in combination with a convolutional neural network to extract long-term health change trends and potential risks; The decision module is used to generate a personalized nursing plan based on the comprehensive health score and provide nursing decision support to medical staff.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for analyzing safety hazards in clinical nursing of neurology according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for analyzing safety hazards in clinical nursing of neurology according to any one of claims 1 to 7 are implemented.

Images