A clinical care system for interventional procedures

Abstract

The present application relates to the technical field of clinical nursing, in particular to a clinical nursing system for interventional surgery, in the present application, through the combination of preoperative medical history, intraoperative real-time physiological data and operation type, depth neural network analysis can accurately identify and predict potential complications, and the nursing plan can be adjusted according to the individual differences of patients, the nursing plan can be adjusted according to the individual differences of patients, the depth neural network can process complex physiological data and identify key physiological data, the long short-term memory network can dynamically adjust the scheme in the nursing process, the change trend of intraoperative data is deeply mined, the physiological index fluctuation is tracked, the abnormality is identified in time and the potential risk is predicted, the nursing decision can be updated synchronously with the physiological state of the patient, the working efficiency of the nursing staff is optimized, and the postoperative recovery effect and comfort of the patient are improved.

Description

Technical Field

[0001] This invention relates to the field of clinical nursing technology, and in particular to a clinical nursing system for interventional surgery. Background Technology

[0002] The field of clinical nursing technology aims to improve patient treatment outcomes, ensure patient safety, and optimize the use of nursing resources. It involves aspects such as patient health monitoring, the development and implementation of nursing plans, the management and optimization of the nursing process, the improvement of patient comfort, and the enhancement of nursing staff work efficiency, thereby promoting rapid patient recovery and reducing medical errors.

[0003] The purpose of a clinical nursing system for interventional surgery is to optimize nursing services during interventional surgery by integrating information technology and nursing processes, ensuring patient safety and comfort during the procedure. It aims to achieve functions such as real-time monitoring, data collection, nursing plan management, and patient information integration, assisting clinical nurses in making timely and accurate nursing decisions, improving nursing efficiency, and enhancing patient surgical safety and recovery outcomes.

[0004] Current technologies lack real-time in-depth analysis and dynamic adjustment of patients' physiological status during surgery. Nursing procedures are fixed and difficult to make effective personalized adjustments based on individual patient differences. They fail to make full use of real-time data during surgery for accurate judgment, resulting in nursing plans lagging behind changes in the patient's condition. They fail to track and adjust physiological indicators in real time during the nursing process, affecting the patient's recovery. This forces clinical nurses to intervene only after complications occur, missing the best intervention opportunity. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing a clinical nursing system for interventional procedures.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a clinical nursing system for interventional surgery includes:

[0007] Complication risk prediction module: Based on the patient's real-time physiological data during the operation, preoperative medical history and operation type, a deep neural network is used to compare and analyze physiological data, identify key physiological data that may cause complications, assess the risk of complications, determine the type of potential complications, and generate complication risk prediction results.

[0008] Personalized nursing care plan development module: Based on the complication risk prediction results and historical case data, combined with physiological data, assess individual patient differences, select appropriate nursing content, and generate personalized nursing care plans;

[0009] Real-time adjustment module for nursing plan: Based on the personalized nursing plan, continuously monitor changes in physiological data, analyze drug response and fluctuations in vital signs, determine whether the personalized nursing plan needs to be adjusted based on real-time physiological data, and generate a dynamic nursing adjustment plan.

[0010] Physiological data time series analysis module: acquires intraoperative physiological data in real time, performs time series analysis, tracks the trend of changes in the patient's physiological indicators, analyzes abnormal fluctuations, predicts potential risks, and generates physiological data time series change results;

[0011] Nursing execution decision module: Based on the time-series changes in the physiological data and the dynamic nursing adjustment plan, the module uses a long short-term memory network, combined with the patient's physiological state, to select appropriate nursing strategies and generate a nursing execution decision plan.

[0012] As a further aspect of the present invention, the complication risk prediction module includes:

[0013] Data comparison submodule: Based on the patient's real-time physiological data during the operation, preoperative medical history and operation type, the module compares the patient's physiological data, uses a deep neural network to collect blood pressure, heart rate and blood oxygen concentration indicators, analyzes the deviation between current values ​​and historical values, evaluates the changes of various physiological data, and generates physiological data comparison results.

[0014] Risk assessment submodule: Based on the physiological data comparison results, assess whether the patient's blood pressure and heart rate exceed the safe range during the operation, analyze the changes in various risk indicators, combine the risk factors in the patient's medical history, calculate the complication risk score, and generate the complication risk assessment results;

[0015] Complication assessment submodule: Based on the complication risk assessment results, combined with the patient's medical history and real-time physiological data obtained during the surgical procedure, analyze the risk factors associated with relevant complications, determine the type of complications the patient will experience, and generate complication risk prediction results.

[0016] As a further aspect of the present invention, the deep neural network adopts the formula:

[0017]

[0018] in: This is the total deviation value. For the current moment, the first Real-time data of physiological data, For the corresponding number in historical data The values ​​of the physiological data, Based on average physiological data from the same patient population, The threshold for physiological data obtained through evaluation by medical experts. For target physiological data values, , , , These are the weighting coefficients for each data point. This is to compare the total number of physiological data involved.

[0019] As a further aspect of the present invention, the personalized care plan formulation module includes:

[0020] Patient Difference Assessment Submodule: Based on the complication risk prediction results and historical case data, extract patients' age, weight and past medical history information, analyze the impact of various physiological data and historical factors on nursing needs, assess individual differences, and generate individual difference assessment results;

[0021] Nursing content selection submodule: Based on the individual difference assessment results, combined with the patient's physiological data, and compared with the expected range of physiological changes during the operation, the patient's recovery is assessed, appropriate nursing measures are selected, and nursing content selection results are generated;

[0022] Nursing plan generation submodule: Based on the nursing content selection results, integrate individual patient differences and appropriate nursing content, formulate detailed nursing plans, specify specific nursing steps, and generate personalized nursing plans;

[0023] The specific nursing steps include vital sign monitoring, wound care, nutritional management, activity guidance and functional training, pain management, psychological and emotional support, complication prevention and monitoring, respiratory function training, medication management, discharge preparation and follow-up care planning.

[0024] As a further aspect of the present invention, the real-time adjustment module for the nursing plan includes:

[0025] Physiological data monitoring submodule: Based on the personalized care plan, continuously collect the current patient's physiological data, monitor changes in physiological data in real time, compare with historical records, and generate physiological data monitoring results;

[0026] Nursing response analysis submodule: Based on the physiological data monitoring results, support vector machine is used to automatically analyze the changes in the patient's physiological data, identify the patterns of drug response and sign fluctuations, construct a time series analysis model, detect the changing trends of the patient's physiological data in real time, and judge the possibility of drug side effects and complication signs by combining the set thresholds, and generate nursing response analysis results.

[0027] Nursing plan adjustment submodule: Based on the nursing response analysis results and combined with real-time physiological data changes, nursing measures are matched according to the patient's health status to generate a dynamic nursing adjustment plan.

[0028] As a further aspect of the present invention, the physiological data time series analysis module includes:

[0029] Data collection submodule: Acquires intraoperative physiological data in real time, organizes it into time series data in chronological order, and generates physiological data collection results;

[0030] Data trend analysis submodule: Based on the physiological data collection results, the module uses time series analysis to identify the evolution pattern of each physiological data item in the time dimension, analyzes the long-term change trend of different physiological indicators, explores potential periodic changes and abnormal patterns, predicts the future direction of physiological changes, and generates physiological data trend analysis results.

[0031] Risk prediction submodule: Based on the trend analysis results of the physiological data, analyze and identify abnormal fluctuations in the physiological data, determine whether there are potential risk factors, predict the patient's physiological abnormalities and complications, and generate the time-series change results of the physiological data.

[0032] As a further aspect of the present invention, the nursing execution decision module includes:

[0033] Data Analysis Submodule: Based on the time-series changes in the physiological data and the dynamic nursing adjustment plan, collect the patient's current physiological data, use a long short-term memory network to classify and organize the various physiological data, calculate the change rate of each physiological data, identify the fluctuation range of each physiological data, and generate data analysis results;

[0034] Nursing plan comparison submodule: Based on the data analysis results, compare the effects of different nursing plans under the same physiological conditions, conduct simulation evaluation of each plan, analyze the impact of each plan on the patient's physiological indicators, and generate nursing plan comparison results;

[0035] Nursing decision generation submodule: Based on the comparison results of the nursing plans, combined with the patient's individual needs and physiological responses, select appropriate nursing measures and generate a nursing execution decision plan.

[0036] As a further aspect of the present invention, the Long Short-Term Memory network adopts the formula:

[0037]

[0038] in: For the first The weighted relative rate of change of the physiological data. For the first The value of the physiological data at the current moment. For the first The value of the physiological data at the previous moment. For the first The population mean of the physiological data, For the first Medical thresholds for physiological data The weighting coefficients for the physiological data at the current moment. The weighting coefficients for physiological data at historical moments. The weighting coefficients for the group average. The weighting coefficients for medical thresholds. These are the normalized weighting coefficients for historical data.

[0039] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0040] 1. In this invention, by combining preoperative medical history, real-time intraoperative physiological data and surgical type for deep neural network analysis, potential complications can be accurately identified and predicted, ensuring that nursing plans can be finely adjusted based on individual patient differences.

[0041] 2. In this invention, deep neural networks can process complex physiological data and identify key physiological data to assess the risk of complications, making real-time data comparison and analysis more accurate and efficient, and ensuring timely assessment and effective management of risks during surgery;

[0042] 3. In this invention, the long short-term memory network is used to dynamically adjust the nursing plan during the nursing process, deeply explore the changing trends of intraoperative data, track the fluctuations of physiological indicators, identify abnormalities in a timely manner and predict potential risks, ensure that nursing decisions can be updated in sync with the patient's physiological state, optimize the work efficiency of nursing staff, and improve the postoperative recovery and comfort of patients. Attached Figure Description

[0043] Figure 1 This is a system flowchart of the present invention;

[0044] Figure 2 This is a schematic diagram of the system framework of the present invention. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0046] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0047] Please see Figure 1 The present invention provides a technical solution: a clinical nursing system for interventional surgery comprising:

[0048] Complication risk prediction module: Based on the patient's real-time physiological data during the operation, preoperative medical history and operation type, a deep neural network is used to compare and analyze physiological data, identify key physiological data that may cause complications, assess the risk of complications, determine the type of potential complications, and generate complication risk prediction results.

[0049] Personalized nursing care plan development module: Based on complication risk prediction results and historical case data, combined with physiological data, assess individual patient differences, select appropriate nursing content, and generate personalized nursing care plans;

[0050] Real-time adjustment module for nursing plans: Based on personalized nursing plans, continuously monitor changes in physiological data, analyze drug responses and fluctuations in vital signs, determine whether personalized nursing plans need to be adjusted based on real-time physiological data, and generate dynamic nursing adjustment plans.

[0051] Physiological data time series analysis module: acquires intraoperative physiological data in real time, performs time series analysis, tracks the trend of changes in the patient's physiological indicators, analyzes abnormal fluctuations, predicts potential risks, and generates physiological data time series change results;

[0052] Nursing execution decision module: Based on the time-series changes in physiological data and dynamic nursing adjustment plans, it uses a long short-term memory network, combined with the patient's physiological state, to select appropriate nursing strategies and generate nursing execution decision plans.

[0053] Please see Figure 2 The complication risk prediction module includes:

[0054] Data comparison submodule: Based on the patient's real-time physiological data during the operation, preoperative medical history and operation type, the module compares the patient's physiological data, uses a deep neural network to collect blood pressure, heart rate and blood oxygen concentration indicators, analyzes the deviation between current values ​​and historical values, evaluates the changes of various physiological data, and generates physiological data comparison results.

[0055] Risk assessment submodule: Based on the comparison results of physiological data, assess whether the patient's blood pressure and heart rate indicators exceed the safe range during the operation, analyze the changes in various risk indicators, combine the risk factors in the patient's medical history, calculate the complication risk score, and generate the complication risk assessment results;

[0056] Complication assessment submodule: Based on the complication risk assessment results, combined with the patient's medical history and real-time physiological data obtained during the surgical procedure, analyze the risk factors associated with relevant complications, determine the type of complications the patient will experience, and generate complication risk prediction results;

[0057] Data Comparison Submodule: Based on the patient's real-time physiological data during surgery, preoperative medical history, and surgical type, this module compares the patient's physiological data. It employs a deep neural network with three hidden layers, each containing 512, 256, and 128 neurons. The ReLU activation function is used. Blood pressure, heart rate, and blood oxygen saturation are input as parameters. The module uses a backpropagation algorithm with gradient descent, a learning rate of 0.001, and a momentum of 0.9. It compares the deviation between the current and historical values, calculates the error using mean squared error, optimizes parameters, evaluates changes in various physiological data, and generates physiological data comparison results.

[0058] Risk assessment submodule: Based on the comparison results of surgical physiological data, assess whether the patient's blood pressure and heart rate exceed the safe range during the operation, analyze the changes in various risk indicators, combine the risk factors in the patient's medical history, calculate the complication risk score, use the linear regression algorithm, solve the parameters by the least squares method, use the patient's blood pressure and heart rate data as independent variables, set the complication risk score as the dependent variable, fit the patient data, analyze whether it exceeds the set blood pressure and heart rate thresholds, and generate the complication risk assessment result;

[0059] Complication Judgment Submodule: Based on the surgical complication risk assessment results, combined with the patient's medical history and real-time physiological monitoring data during the operation, it analyzes the risk factors associated with related complications, uses the K-nearest neighbor algorithm with K set to 5, calculates similarity using Euclidean distance, selects a simple weighted average method for classification, classifies the patient's risk factors, calculates the similarity of each patient with known cases, determines the type of complications the patient will experience, and generates complication risk prediction results.

[0060] Please see Figure 2 Deep neural networks use the following formula:

[0061]

[0062] in: This is the total deviation value. For the current moment, the first Real-time data of physiological data, For the corresponding number in historical data The values ​​of the physiological data, Based on average physiological data from the same patient population, The threshold for physiological data obtained through evaluation by medical experts. For target physiological data values, , , , These are the weighting coefficients for each data point. To compare the total number of physiological data involved;

[0063] Execution process: First, collect the current patient's physiological data. and historical data The data is compared to historical data to calculate the deviation between current physiological data and historical data, and the current data is compared with the average data based on the same patient population. The deviation of the patient's physiological data is compared and assessed. Next, the current data is compared with the physiological data thresholds assessed by medical experts. Compare the data to ensure the parameter values ​​are within the normal range, and calculate the current data and target physiological data values. To ensure that the patient's physiological state meets the expected goals, all deviation values ​​are determined according to their respective weighting coefficients. , , , Perform a weighted summation to obtain the total deviation value. It is used to assess whether there are any abnormalities in the patient's current physiological state and to provide clinical decision support.

[0064] Please see Figure 2 The personalized care plan development module includes:

[0065] Patient Difference Assessment Submodule: Based on complication risk prediction results and historical case data, extract patients’ age, weight and past medical history information, analyze the impact of various physiological data and historical factors on nursing needs, assess individual differences, and generate individual difference assessment results;

[0066] Nursing content selection submodule: Based on the individual difference assessment results, combined with the patient's physiological data, and compared with the expected range of physiological changes during the operation, the patient's recovery is assessed, appropriate nursing measures are selected, and nursing content selection results are generated;

[0067] Nursing plan generation submodule: Based on the nursing content selection results, integrate individual patient differences and appropriate nursing content, formulate detailed nursing plans, specify specific nursing steps, and generate personalized nursing plans;

[0068] Specific nursing steps include vital sign monitoring, wound care, nutritional management, activity guidance and functional training, pain management, psychological and emotional support, complication prevention and monitoring, respiratory function training, medication management, discharge preparation and follow-up care planning.

[0069] The patient difference assessment submodule extracts patients' age, weight, and past medical history information based on complication risk prediction results and historical case data. It analyzes the impact of various physiological and historical factors on nursing needs, and uses a decision tree algorithm, specifically the CART algorithm. The maximum tree depth is set to 10, the minimum number of sample splits is 2, and the minimum number of leaf node samples is 1. The splits are performed using the Gini coefficient, and the CART algorithm is used to construct the decision tree model. The importance of each factor in predicting nursing needs is analyzed, and individual differences are assessed to generate individual difference assessment results.

[0070] Nursing content selection submodule: Based on the individual difference assessment results and combined with the patient's physiological data, the logistic regression algorithm is used to assess the patient's recovery status, select appropriate nursing measures, and use the logistic regression algorithm with a regularization coefficient of 0.01 to fit the patient's physiological data and recovery status through maximum likelihood estimation to generate the selection probability of nursing measures, select appropriate nursing steps, and generate nursing content selection results.

[0071] The nursing plan generation submodule: Based on the nursing content selection results, it integrates individual patient differences with suitable nursing content, formulates detailed nursing plans, and specifies specific nursing steps. A linear programming algorithm is used, optimized using the Simplex method. A variable set xi is constructed to represent whether the i-th nursing measure is selected; xi is 0 or 1. Nursing resource constraints include a daily available nursing time not exceeding 480 minutes, a maximum daily working time for a single nursing staff member not exceeding 240 minutes, and a maximum of 3 nursing measures to be implemented simultaneously. The constraint expression is in the form Ax ≤ b, where A is the nursing resource occupancy matrix, b is the resource upper limit vector, and resource items include time, personnel, average workload per person, and medication frequency. Each element aij in the matrix represents the amount of resource occupancy of the j-th item by the i-th nursing measure. During the linear programming solution process, the value of xi is continuously adjusted according to resource constraints, and the combination of nursing measures that satisfies all constraints is selected to generate a personalized nursing plan.

[0072] Please see Figure 2 The real-time adjustment module for nursing plans includes:

[0073] Physiological data monitoring submodule: Based on personalized care plans, it continuously collects the current physiological data of patients, monitors changes in physiological data in real time, compares with historical records, and generates physiological data monitoring results;

[0074] Nursing response analysis submodule: Based on physiological data monitoring results, it uses support vector machine to automatically analyze changes in patients' physiological data, identify the patterns of drug response and sign fluctuations, construct a time series analysis model, detect the changing trends of patients' physiological data in real time, and combine the set thresholds to judge the possibility of drug side effects and complication signs, and generate nursing response analysis results.

[0075] Nursing plan adjustment submodule: Based on the nursing response analysis results and combined with real-time physiological data changes, it matches nursing measures according to the patient's health status and generates a dynamic nursing adjustment plan;

[0076] Physiological data monitoring submodule: Based on personalized care plans, it continuously collects patients' physiological data, monitors data changes in real time, compares with historical records, uses the Kalman filtering algorithm, sets the process noise covariance to 0.1 and the measurement noise covariance to 0.05, calculates the weighted average of real-time data through Kalman gain, smooths the noise in physiological data, and generates physiological data monitoring results.

[0077] The nursing response analysis submodule analyzes patients' responses to medications and fluctuations in vital signs based on physiological data monitoring results. It determines whether there are signs of drug side effects and complications. The module uses a support vector machine algorithm with a radial basis function kernel, a C value of 1, and a gamma value of 0.5. Cross-validation (K=5) is used to optimize the model parameters. The support vector machine is used to classify patients' physiological data into categories including normal responses, signs of side effects, and risk of complications, generating nursing response analysis results.

[0078] The nursing plan adjustment submodule: Based on the nursing response analysis results and combined with real-time physiological data changes, it selects appropriate nursing measures, adopts a greedy algorithm, sets the objective function as minimizing nursing resources, and sets the constraints as the range of changes in the patient's physiological indicators and the availability of nursing resources. It selects the optimal nursing measures and generates a dynamic nursing adjustment plan.

[0079] Please see Figure 2 The physiological data time series analysis module includes:

[0080] Data collection submodule: Acquires intraoperative physiological data in real time, organizes it into time series data in chronological order, and generates physiological data collection results;

[0081] Data Trend Analysis Submodule: Based on the physiological data collection results, this module uses time series analysis to identify the evolution pattern of each physiological data point over time, analyze the long-term trends of different physiological indicators, uncover potential periodic changes and abnormal patterns, predict future physiological changes, and generate physiological data trend analysis results.

[0082] Risk prediction submodule: Based on the results of physiological data trend analysis, analyze and identify abnormal fluctuations in physiological data, determine whether there are potential risk factors, predict patients' physiological abnormalities and complications, and generate physiological data time series change results;

[0083] Data collection submodule: Based on the physiological data collection results, the data is organized into time series data in chronological order. SQL queries are used to select all physiological data fields during the operation using the SELECT statement, with the time range limited to the start of the operation to the end of the operation. The data is then sorted in ascending order by the timestamp field to generate the physiological data collection results.

[0084] Data Trend Analysis Submodule: Based on the physiological data collection results, time series analysis is performed on various physiological data to identify the trend of each indicator over time, detect fluctuations in the data, determine whether abnormal changes occur, adopt an autoregressive moving average model, set the ARIMA(1,1,1) model, use the AIC criterion to optimize the model parameters, analyze the time series data of each physiological indicator, and generate physiological data trend analysis results.

[0085] Risk prediction submodule: Based on the trend analysis results of physiological data, analyze and identify abnormal fluctuations in physiological data, determine whether there are potential risk factors, predict possible physiological changes in patients, use an anomaly detection algorithm, set the number of trees to 100 and the maximum depth to 10, calculate the anomaly score for each data point, and based on the anomaly score, if the score is ≥0.65, it is judged as a potential risk, and continuous high-risk scores are marked as abnormal periods, generating the time series change results of physiological data.

[0086] Please see Figure 2 The nursing execution decision module includes:

[0087] Data Analysis Submodule: Based on the time-series changes in physiological data and dynamic nursing adjustment plans, it collects the patient's current physiological data, uses a long short-term memory network to classify and organize various physiological data, calculates the change rate of various physiological data, identifies the fluctuation range of various physiological data, and generates data analysis results;

[0088] Nursing Plan Comparison Submodule: Based on data analysis results, compare the effects of different nursing plans under the same physiological conditions, conduct simulation evaluations of each plan, analyze the impact of each plan on the patient's physiological indicators, and generate nursing plan comparison results;

[0089] Nursing decision generation submodule: Based on the comparison results of nursing options, combined with the patient's individual needs and physiological responses, select appropriate nursing measures and generate nursing execution decision plans;

[0090] Data Analysis Submodule: Based on the time-series changes in physiological data and dynamic nursing adjustment plans, this module collects the patient's current physiological data, employs a Long Short-Term Memory (LSTM) network, sets the input layer dimension to 100, the hidden layer number to 128, and the learning rate to 0.001, and trains it using the Adam optimizer. It then categorizes and organizes the data, calculates the relative changes between different physiological data, performs normalization through time-series data processing, identifies the fluctuation range of each data point, and generates data analysis results.

[0091] Nursing Plan Comparison Submodule: Based on data analysis results, compare the effects of different nursing plans under the same physiological conditions, conduct simulation evaluation of each plan, use Monte Carlo simulation algorithm, set the number of simulations to 1000, simulate the execution effect of each nursing plan, use normal distribution to model the influencing factors of each nursing plan, use mean and standard deviation to evaluate the impact of each nursing plan on the patient's physiological indicators, and generate nursing plan comparison results.

[0092] Nursing decision generation submodule: Based on the comparison results of nursing options, combined with the patient's individual needs and physiological responses, the module selects appropriate nursing measures. It uses a genetic algorithm with a population size of 50, an iteration count of 100, a crossover probability of 0.8, and a mutation probability of 0.1 to evaluate the fitness of each nursing option. The optimal nursing option is selected through a roulette wheel selection method, generating a nursing execution decision plan.

[0093] Please see Figure 2 Long Short-Term Memory (LSTM) networks use the following formula:

[0094]

[0095] in: For the first The weighted relative rate of change of the physiological data. For the first The value of the physiological data at the current moment. For the first The value of the physiological data at the previous moment. For the first The population mean of the physiological data, For the first Medical thresholds for physiological data The weighting coefficients for the physiological data at the current moment. The weighting coefficients for physiological data at historical moments. The weighting coefficients for the group average. The weighting coefficients for medical thresholds. These are the normalized weighting coefficients for historical data;

[0096] Execution process: First, collect the current patient's physiological data. Simultaneously acquire physiological data from the previous moment. Then use the group average and medical threshold To compare the differences between the patient's current data and the normal range and medical standards of the same group, weighting coefficients were assigned to current physiological data, historical data, group average, and medical thresholds. , , , To control the degree of influence of each data item, the weighted difference of each data item is calculated, the absolute values ​​of the differences are summed, and finally the result is divided by the historical data. And multiply by the normalized weighting factor The weighted relative rate of change was obtained. It can more accurately assess changes in the patient's physiological state, providing clinical nurses with a basis for dynamic nursing adjustments.

[0097] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A clinical nursing system for interventional surgery, characterized in that, The system includes: Complication risk prediction module: Based on the patient's real-time physiological data during the operation, preoperative medical history and operation type, a deep neural network is used to compare and analyze physiological data, identify key physiological data that may cause complications, assess the risk of complications, determine the type of potential complications, and generate complication risk prediction results. Personalized nursing care plan development module: Based on the complication risk prediction results and historical case data, combined with physiological data, assess individual patient differences, select appropriate nursing content, and generate personalized nursing care plans; Real-time adjustment module for nursing plan: Based on the personalized nursing plan, continuously monitor changes in physiological data, analyze drug response and fluctuations in vital signs, determine whether the personalized nursing plan needs to be adjusted based on real-time physiological data, and generate a dynamic nursing adjustment plan. Physiological data time series analysis module: acquires intraoperative physiological data in real time, performs time series analysis, tracks the trend of changes in the patient's physiological indicators, analyzes abnormal fluctuations, predicts potential risks, and generates physiological data time series change results; Nursing execution decision module: Based on the time-series changes in the physiological data and the dynamic nursing adjustment plan, a long short-term memory network is used to select appropriate nursing strategies and generate a nursing execution decision plan in combination with the patient's physiological state. The nursing execution decision module includes: Data Analysis Submodule: Based on the time-series changes in the physiological data and the dynamic nursing adjustment plan, collect the patient's current physiological data, use a long short-term memory network to classify and organize the various physiological data, calculate the change rate of each physiological data, identify the fluctuation range of each physiological data, and generate data analysis results; Nursing plan comparison submodule: Based on the data analysis results, compare the effects of different nursing plans under the same physiological conditions, conduct simulation evaluation of each plan, analyze the impact of each plan on the patient's physiological indicators, and generate nursing plan comparison results; Nursing decision generation submodule: Based on the comparison results of the nursing plans, combined with the patient's individual needs and physiological responses, select appropriate nursing measures and generate a nursing execution decision plan; The physiological data time series analysis module includes: Data collection submodule: Acquires intraoperative physiological data in real time, organizes it into time series data in chronological order, and generates physiological data collection results; Data trend analysis submodule: Based on the physiological data collection results, the module uses time series analysis to identify the evolution pattern of each physiological data item in the time dimension, analyzes the long-term change trend of different physiological indicators, explores potential periodic changes and abnormal patterns, predicts the future direction of physiological changes, and generates physiological data trend analysis results. Risk prediction submodule: Based on the trend analysis results of the physiological data, analyze and identify abnormal fluctuations in the physiological data, determine whether there are potential risk factors, predict the patient's physiological abnormalities and complications, and generate the time-series change results of the physiological data; The Long Short-Term Memory (LSTM) network uses the following formula: ； in: For the first The weighted relative rate of change of the physiological data. For the first The value of the physiological data at the current moment. For the first The value of the physiological data at the previous moment. For the first The population mean of the physiological data, For the first Medical thresholds for physiological data The weighting coefficients for the physiological data at the current moment. The weighting coefficients for physiological data at historical moments. The weighting coefficients for the group average. The weighting coefficients for medical thresholds. These are the normalized weighting coefficients for historical data.

2. The clinical nursing system for interventional surgery according to claim 1, characterized in that, The complication risk prediction module includes: Data comparison submodule: Based on the patient's real-time physiological data during the operation, preoperative medical history and operation type, the module compares the patient's physiological data, uses a deep neural network to collect blood pressure, heart rate and blood oxygen concentration indicators, analyzes the deviation between current values ​​and historical values, evaluates the changes of various physiological data, and generates physiological data comparison results. Risk assessment submodule: Based on the physiological data comparison results, assess whether the patient's blood pressure and heart rate exceed the safe range during the operation, analyze the changes in various risk indicators, combine the risk factors in the patient's medical history, calculate the complication risk score, and generate the complication risk assessment results; Complication assessment submodule: Based on the complication risk assessment results, combined with the patient's medical history and real-time physiological data obtained during the surgical procedure, analyze the risk factors associated with relevant complications, determine the type of complications the patient will experience, and generate complication risk prediction results.

3. The clinical nursing system for interventional surgery according to claim 2, characterized in that, The deep neural network uses the following formula: ； in: This is the total deviation value. For the current moment Real-time data of physiological data, For the corresponding number in historical data The values ​​of the physiological data, Based on average physiological data from the same patient population, The threshold for physiological data obtained through evaluation by medical experts. For target physiological data values, , , , These are the weighting coefficients for each data point. This is to compare the total number of physiological data involved.

4. The clinical nursing system for interventional surgery according to claim 1, characterized in that, The personalized care plan development module includes: Patient Difference Assessment Submodule: Based on the complication risk prediction results and historical case data, extract patients' age, weight and past medical history information, analyze the impact of various physiological data and historical factors on nursing needs, assess individual differences, and generate individual difference assessment results; Nursing content selection submodule: Based on the individual difference assessment results, combined with the patient's physiological data, and compared with the expected range of physiological changes during the operation, the patient's recovery is assessed, appropriate nursing measures are selected, and nursing content selection results are generated; Nursing plan generation submodule: Based on the nursing content selection results, integrate individual patient differences and appropriate nursing content, formulate detailed nursing plans, specify specific nursing steps, and generate personalized nursing plans; The specific nursing steps include vital sign monitoring, wound care, nutritional management, activity guidance and functional training, pain management, psychological and emotional support, complication prevention and monitoring, respiratory function training, medication management, discharge preparation and follow-up care planning.

5. The clinical nursing system for interventional surgery according to claim 1, characterized in that, The real-time adjustment module for the nursing plan includes: Physiological data monitoring submodule: Based on the personalized care plan, continuously collect the current patient's physiological data, monitor changes in physiological data in real time, compare with historical records, and generate physiological data monitoring results; Nursing response analysis submodule: Based on the physiological data monitoring results, support vector machine is used to automatically analyze the changes in the patient's physiological data, identify the patterns of drug response and sign fluctuations, construct a time series analysis model, detect the changing trends of the patient's physiological data in real time, and judge the possibility of drug side effects and complication signs by combining the set thresholds, and generate nursing response analysis results. Nursing plan adjustment submodule: Based on the nursing response analysis results and combined with real-time physiological data changes, nursing measures are matched according to the patient's health status to generate a dynamic nursing adjustment plan.

Images