Intelligent collaborative nursing system for patients with cardiovascular intervention surgery in PACU

Abstract

The application discloses an intelligent collaborative nursing system for patients of cardiovascular intervention surgery in a PACU, relates to the technical field of postoperative nursing, and solves the technical problem of low timeliness in the anesthesia recovery stage caused by the lack of real-time change processing of the patient in the prior art; the system first synchronously collects multi-mode data of the patient and continuously collects real-time dynamic monitoring data of the postoperative patient; the data are respectively quantitatively assigned to obtain a static risk score, a dynamic correction score and a nursing risk score; the static risk score and the nursing risk score are normalized and coupled, and grade weight fusion is performed to obtain a static risk estimate; the baseline is denoised and corrected by combining the deviation disturbance quantity of the dynamic correction score, and an accurate postoperative combined risk estimate of the patient is output; the system divides the postoperative risk level according to the combined risk estimate, presets standardized PACU nursing collaborative operations adapted to different risk levels as required, and arranges and adaptively adjusts and optimizes the preset operations in time sequence.

Description

Technical Field

[0001] This invention belongs to the field of postoperative care technology, specifically an intelligent collaborative care system for cardiovascular interventional surgery patients in PACU. Background Technology

[0002] Cardiovascular interventional surgery is a highly precise and high-risk clinical procedure. Patients require postoperative care in the PACU (Postoperative Care Unit) for resuscitation and observation, and are highly susceptible to sudden complications such as arrhythmias, puncture site bleeding, blood pressure fluctuations, and contrast-induced kidney injury. Therefore, accurate postoperative risk assessment and dynamically adaptive nursing interventions are crucial for ensuring patient resuscitation safety. Currently, clinical PACU nursing relies heavily on the subjective experience of medical staff to determine risks and configure nursing procedures. Risk assessment often depends solely on some intraoperative vital signs, failing to achieve multi-source fusion analysis of static baseline surgical data, intraoperative dynamic fluctuation data, and data from the surgical adaptation nursing plan. This results in a one-sided risk assessment dimension, low quantitative accuracy, and an inability to truly cover the patient's inherent disease risks and real-time postoperative fluctuation risks.

[0003] Patent application number 2025111329203 discloses a risk prediction system for postoperative care of cardiovascular intervention based on physiological feedback. This invention first acquires historical patient physiological data and postoperative care risk values ​​and groups them; further analyzes the clustering of patient risk values ​​and combines them to obtain initial typicality; further analyzes the reference weights of other groups for various physiological data of the target group, and uses this to analyze the correlation between the physiological data of other groups and the maximum initial typicality within other groups, obtaining the predictive importance of various physiological data of the target group; further obtains the final typicality; finally, based on the final typicality of patients within each group, it integrates the risk value to obtain the risk optimization value and constructs a predictive decision tree, improving the representativeness and individualized predictive ability of the decision tree. However, when obtaining operational decisions for postoperative care, this invention requires analysis of the patient's overall data. Because there are many common feature points for the same surgery, individual calculations would increase the amount of data to be processed, resulting in slow processing speed. Furthermore, during the anesthesia recovery phase, timely changes in operation are required in response to real-time changes in the patient. Summary of the Invention

[0004] The present invention aims to at least solve one of the technical problems existing in the prior art; to this end, the present invention proposes an intelligent collaborative nursing system for cardiovascular interventional patients in PACU, which is used to solve the technical problem of low timeliness of treatment during the anesthesia recovery phase due to the lack of processing of real-time changes in patients in the prior art.

[0005] To achieve the above objectives, a first aspect of the present invention provides an intelligent collaborative care system for patients undergoing cardiovascular interventional procedures within a PACU, comprising: Acquire surgical data during cardiovascular interventional procedures and nursing data after surgery, and collect dynamic monitoring data of patients in real time; By combining surgical and nursing data, a risk assessment is conducted on the patient's condition after cardiovascular interventional surgery, and based on the results of the risk assessment, collaborative nursing procedures that the patient needs are planned. The preset collaborative operations are dynamically optimized and adjusted based on dynamic monitoring data, and the results are recorded.

[0006] Preferably, the risk assessment of the patient's condition after cardiovascular interventional surgery, combining surgical data and nursing data, includes: Extract surgical and nursing data; the surgical data includes intraoperative static data and dynamic baseline data; the nursing data consists of nursing plans developed based on the type of surgery. Intraoperative static data are quantified to obtain static risk scores, dynamic baseline data are quantified to obtain dynamic correction scores, and nursing data are quantified to obtain nursing risk scores. By combining the static risk score and the nursing risk score, the static risk estimate of the patient's surgery is obtained. The dynamic correction score is used as the impact data of the static risk estimate to correct it, resulting in the combined risk estimate.

[0007] Preferably, the step of combining and superimposing static risk scores and nursing risk scores to obtain a static risk assessment of the patient's surgery includes: The static risk score and the nursing risk score are normalized, and the normalized static risk score and the nursing risk score are coupled and merged to obtain the co-coupled risk value. The risk prediction level is obtained by comparing the collaborative coupling risk value with the preset risk threshold. Based on the risk prediction level, corresponding level weight coefficients are set for the static risk score and the nursing risk score. The static risk estimate is obtained by combining the synergistic coupling risk value, static risk score, and nursing risk score with their respective grade weight coefficients.

[0008] Preferably, the step of adjusting the dynamic correction score as the impact data of the static risk assessment to obtain the combined risk assessment includes: The dynamic correction score is normalized, and the normalized dynamic correction score is compared with the standard dynamic correction value in the database to obtain the deviation disturbance. Using static risk valuation as a fixed risk baseline, the combined risk valuation is obtained by superimposing and correcting the fixed risk baseline through deviation disturbance.

[0009] Understandably, when superimposing and correcting a fixed risk baseline using deviation disturbances, it is necessary to denoise the disturbances to avoid the impact of some abnormal disturbances on the correction results.

[0010] Preferably, the pre-setting of collaborative operations for patient care based on the results of risk assessment includes: The postoperative risk level of cardiovascular interventional surgery will be determined by combining risk assessment. For different postoperative risk levels, corresponding standardized and specialized PACU nursing basic operations and key special operations will be preset to form a graded nursing operation library. All preset nursing collaborative operations in the hierarchical nursing operation library are arranged in a time-sequential and process-oriented manner, and adaptive replacement logic is added to the steps of all preset nursing collaborative operations. When marked correction data is detected, the preset nursing collaborative operations are adaptively adjusted. Among them, the correction data is the data when the surgical data and nursing data are updated, and the marked correction data is the data that has a significant impact on the preset nursing collaborative operations.

[0011] Preferably, the step of dynamically optimizing and adjusting the preset collaborative operation based on dynamic monitoring data and recording it includes: The operation optimization model is obtained by training a neural network model using standard monitoring data, preset collaborative operations, and corresponding adjustment operation data from the database as training data. By incorporating real-time dynamic monitoring data into the operation optimization model, the model updates patient risk assessment and risk level in real time, and simultaneously verifies the adaptability of the currently implemented preset nursing collaborative operation plan, and performs hierarchical adaptive optimization and adjustment of the preset nursing collaborative operation.

[0012] Preferably, the step of training a neural network model to obtain an operation optimization model by using standard monitoring data from the database, preset collaborative operations, and corresponding adjustment operation data as training data includes: Select a model framework and deep learning algorithm from the artificial intelligence library; build a model based on the model framework and deep learning algorithm to obtain the initial model; Obtain the standard dataset; the standard dataset includes standard input data consistent with the content attributes of the state analysis sequence, and standard output data consistent with the content attributes of the state labels; The standard dataset is divided into a training set, a validation set, and a test set according to a set ratio; the initial model is trained using the training set; the internal structure and parameters of the initial model are adjusted using the validation set; and the trained initial model is tested using the test set to obtain test metrics. Obtain the indicator threshold; compare the test indicator with the indicator threshold; if all test indicators are greater than the indicator threshold, mark the initial model as the operation optimization model; otherwise, rebuild and retrain the initial model.

[0013] Preferably, the synchronization performs an adaptability check on the currently executing preset nursing collaborative operation plan, including: The system inputs real-time dynamic monitoring data into the risk assessment model, updates the patient's combined risk estimate and risk level in real time, integrates the combined risk estimate and risk level to obtain the risk status, and performs an adaptability check on the currently executed preset nursing collaborative operation plan based on the updated risk status. It determines from multiple dimensions whether the current nursing plan matches the patient's real-time condition. When there is an index deviation or risk level rise or fall, a dynamic optimization and adjustment mechanism is triggered.

[0014] Preferably, the hierarchical adaptive optimization adjustment of the preset nursing collaborative operations includes: Pre-defined collaborative nursing operations are stratified based on the principles of operation time period and operation configuration to obtain stratified nursing operations. The operation time period and operation configuration are used as tags and corresponding stratified nursing operations. The system detects the labeled correction data, matches the configuration correction data according to the labels of the labeled correction data, matches and compares the configuration correction data with the preset nursing collaborative operation to generate operation replacement commands, identifies the operation replacement commands, and performs hierarchical operation replacement optimization adjustments.

[0015] Preferably, the step of matching and comparing the configuration correction data with preset nursing collaborative operations to generate operation replacement commands includes: Identify the labels corresponding to the corrected data, match them with the corresponding standard nursing configurations, and compare them with the stratified nursing operations to obtain the configuration correction data that needs to be corrected; Match the nursing collaboration operations corresponding to the target stratum of the configuration correction data, and compare and verify the compatibility to determine whether a replacement operation is needed; if yes, generate an operation replacement instruction; otherwise, identify the next time period within the stratum.

[0016] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention collects intraoperative static data, dynamic baseline data, and specialized nursing care plan data matching the surgical type from patients undergoing cardiovascular interventional surgery, and continuously collects real-time dynamic monitoring data of patients after surgery. By quantifying and assigning values ​​to these three core data types, static risk scores, dynamic correction scores, and nursing risk scores are obtained. Static risk estimates are obtained by normalizing and coupling the static and nursing risk scores and fusing them with hierarchical weights. The baseline is then denoised and corrected by combining the deviation disturbance of the dynamic correction score, resulting in an accurate postoperative risk estimate for the patient. The system classifies postoperative risk levels based on the combined risk estimate, presets standardized PACU nursing collaborative operations to suit different risk levels as needed, and performs temporal arrangement and adaptive logic embedding of the preset operations. Simultaneously, relying on a trained neural network operation optimization model, the system verifies the adaptability of the current nursing plan to the patient's dynamic condition in real time. Based on the patient's real-time vital sign fluctuations and risk changes, the preset nursing collaborative operations are dynamically optimized and adjusted, and the entire process is recorded and stored. This invention breaks away from the traditional experience-based, homogenized nursing model, achieving a panoramic integrated assessment of surgical risk, nursing risk, and real-time condition.

[0017] 2. This invention introduces a time-based and configuration-based hierarchical replacement optimization mechanism. The system divides all preset nursing collaborative operations into two-dimensional hierarchical divisions according to operation time periods and configurations, and binds unique time-series and attribute tags to each hierarchical nursing operation, constructing a structured and independent hierarchical nursing unit library. During dynamic optimization, the system can accurately locate the target hierarchical unit that needs adjustment based on the patient's real-time resuscitation stage and dynamic risk level through tags. It independently judges and replaces only the local hierarchical nursing operations with insufficient adaptability, without changing the entire nursing plan. This achieves a refined optimization logic of immediate replacement of incompatible layers and stable retention of compatible layers. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram illustrating the working principle of the present invention.

[0020] Figure 2 This is a flowchart illustrating the process of an embodiment of the present invention. Detailed Implementation

[0021] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Please see Figure 1 The first aspect of the present invention provides an intelligent collaborative nursing system for patients undergoing cardiovascular interventional surgery in a PACU, comprising: Acquire surgical data during cardiovascular interventional procedures and nursing data after surgery, and collect dynamic monitoring data of patients in real time; By combining surgical and nursing data, a risk assessment is conducted on the patient's condition after cardiovascular interventional surgery, and based on the results of the risk assessment, collaborative nursing procedures that the patient needs are planned. The preset collaborative operations are dynamically optimized and adjusted based on dynamic monitoring data, and the results are recorded.

[0023] Specifically, in combination Figure 1 and Figure 2 The main purpose of this embodiment is to formulate a nursing plan for patients undergoing cardiovascular interventional surgery in the post-anesthesia monitoring and treatment room, acquire surgical data during the cardiovascular interventional surgery, nursing data during the post-operative PACU care stage, and continuously collect dynamic monitoring data of the patient's entire resuscitation process in real time, providing complete data support for subsequent risk assessment and intelligent nursing control. Among them, the surgical data includes intraoperative static data and dynamic baseline data. Intraoperative static data specifically refers to data during the cardiovascular interventional surgery, including but not limited to fixed attribute data such as underlying diseases, cardiac function status, surgical puncture method, and surgical complexity level. Dynamic baseline data specifically refers to data on surgical monitoring instruments during the surgery, including but not limited to blood pressure, heart rate, blood loss, contrast agent dosage, and intraoperative medication fluctuation data. The nursing data is the data of a specialized nursing plan personalized according to the specific type of cardiovascular interventional surgery for the patient, including nursing operations, nursing time periods, and nursing configurations. The nursing configurations specifically refer to nursing personnel and individual operations.

[0024] Differential quantitative assignments were performed on the collected data. Weighted quantification was performed on various fixed high-risk factors of the intraoperative static data to obtain static risk scores that characterize the patient's inherent underlying condition. Real-time fluctuation indicators of the intraoperative dynamic baseline data were graded and assigned according to the degree of abnormality and duration to obtain dynamic correction scores. The monitoring intensity, intervention severity, and risk control level of the nursing plan were standardized and quantified to obtain nursing risk scores.

[0025] The obtained static risk score, nursing risk score, and dynamic correction score are normalized respectively. The normalized static risk score and nursing risk score are then coupled and merged to generate a co-coupled risk value that includes both the patient's inherent surgical risk and nursing management risk. Next, the system compares the co-coupled risk value with internally preset multi-level risk thresholds to complete the preliminary risk level prediction and obtain the risk prediction level. Finally, based on the risk prediction level, the static risk score and nursing risk score are adaptively matched with corresponding differentiated level weight coefficients. Under high risk levels, the inherent surgical risk weight is emphasized, while under low risk levels, the nursing management risk weight is also considered. The co-coupled risk value, the two types of basic scores, and the corresponding level weight coefficients are weighted and fused to obtain a stable and accurate static risk estimate of the patient's surgery.

[0026] It should be noted that the weighting coefficients for the static risk score and the nursing risk score are different in this embodiment. Specifically, the weighting coefficients for the static risk score and the nursing risk score are set differently depending on the risk prediction level obtained. For example, the risk prediction levels obtained in this embodiment include low risk, medium risk, and high risk. The weighting coefficient for the static risk score is 0.3 for low risk and 0.2 for nursing risk; the weighting coefficient for the static risk score is 0.6 for medium risk and 0.5 for nursing risk; and the weighting coefficient for the static risk score is 0.8 for high risk and 0.7 for nursing risk.

[0027] The normalized dynamic correction score is precisely compared with the standard dynamic correction value of the same type of surgery pre-stored in the system database to calculate the deviation disturbance of the patient's current intraoperative state relative to the standard state. This deviation disturbance represents the degree of negative impact of the patient's real-time intraoperative condition fluctuations on the postoperative recovery risk. The system uses the static risk estimate obtained above as the patient's fixed postoperative risk baseline, and uses the deviation disturbance as a real-time influencing factor to superimpose and correct the fixed risk baseline. At the same time, this embodiment adds an intelligent noise reduction processing mechanism during the correction process, and uses a time-series filtering algorithm to filter out invalid disturbance data such as instantaneous changes in equipment, single accidental fluctuations, and acquisition noise, so as to avoid non-pathological abnormal disturbances interfering with the risk assessment results.

[0028] In this embodiment, static risk assessment is used as a fixed risk baseline. The specific function of the fixed risk baseline is to treat the impact of routine procedures during surgery as a fixed impact value, and to treat the dynamic correction score caused by different patient conditions as a variable affecting this fixed value. The data processing process is divided into dynamic and static, which can better reduce the amount of data to be processed.

[0029] It should be added that when performing overlay correction, a critical threshold needs to be set for the impact of the disturbance on the fixed risk baseline. When the disturbance is less than the critical threshold, it means that the disturbance has a small impact on the fixed risk baseline and can be ignored. Alternatively, a larger risk threshold can be set. When the disturbance is greater than the risk threshold, it means that the data at this time may be abnormal data, or the disturbance has a huge impact on the fixed risk baseline, causing the fixed risk baseline to no longer be applicable to the current situation, and an emergency update or reporting of the situation is required.

[0030] The postoperative risk level of cardiovascular interventional surgery will be determined by combining risk assessment. For different postoperative risk levels, corresponding standardized and specialized PACU nursing basic operations and key special operations will be preset to form a hierarchical nursing operation library. The operation library includes different operations corresponding to different risk levels. For example, routine basic monitoring operations will be preset for low-risk patients, enhanced key monitoring and special physical examination operations will be preset for intermediate-risk patients, and special intervention for critical care, multi-role collaborative monitoring, and pre-complication prevention and control operations will be preset for high-risk patients.

[0031] The system unifies and streamlines all preset collaborative operations within the hierarchical nursing operation library, clarifying the execution sequence, frequency, review nodes, and multi-role division of labor for each operation. Simultaneously, it embeds adaptive replacement logic into all preset operations. The system monitors the updates and changes in surgical and nursing data in real time, marking key data changes that have a significant impact on nursing operations. Once a marked correction data update is detected, the system automatically triggers the adaptive pre-adjustment of the preset nursing collaborative operations, providing a pre-adaptation basis for subsequent dynamic optimization and adjustment.

[0032] It should be noted that the corrected data specifically refers to key core update data that can directly change the patient's risk level, intervention nursing operation standards, and adjust the monitoring level. Specifically, it includes two categories: surgical data updates and nursing data updates. Once the marked corrected data update is detected, the adaptive pre-adjustment of the preset nursing collaborative operation is automatically triggered, providing a preliminary adaptation basis for subsequent dynamic optimization and adjustment.

[0033] Select a model framework and deep learning algorithm from the artificial intelligence library; build a model based on the model framework and deep learning algorithm to obtain an initial model; obtain a standard dataset; the standard dataset includes standard input data consistent with the content attributes of the state analysis sequence and standard output data consistent with the content attributes of the state label; divide the standard dataset into a training set, a validation set, and a test set according to a set ratio; train the initial model using the training set; adjust the internal structure and parameters of the initial model using the validation set; test the trained initial model using the test set to obtain test metrics; obtain metric thresholds; compare the test metrics with the metric thresholds; when all test metrics are greater than the metric thresholds, mark the initial model as an operational optimization model; otherwise, rebuild and retrain the initial model.

[0034] The system continuously feeds real-time dynamic monitoring data into the trained operation optimization model. On the one hand, it iteratively updates the patient's combined risk assessment and postoperative risk level in real time. On the other hand, it performs a comprehensive adaptability check on the currently executed preset nursing collaborative operation plan. It checks the matching degree between the plan and the real-time condition from multiple dimensions such as monitoring frequency, specific operation content, multi-role collaboration level, and clinical intervention intensity. When the matching degree is detected to exceed the set threshold, that is, when it is not applicable to the current risk baseline, the system immediately triggers the dynamic optimization adjustment mechanism, starts the hierarchical adaptive optimization logic, and intelligently adjusts the preset nursing collaborative operation.

[0035] For all pre-set nursing collaborative operations with completed time-series arrangement and division of labor, the system is divided into layers according to the dual dimensions of operation time period and operation configuration, completing standardized layered decomposition. The operation time period is the execution time node and duration of each nursing collaborative operation in the PACU resuscitation process, and the operation configuration is the monitoring level, operation type, intervention intensity, and multi-role collaboration configuration parameters corresponding to each nursing collaborative operation. Based on the above two principles, the system decomposes the overall nursing collaborative operation into multiple independent and functionally defined layered nursing operation units, resulting in layered nursing operations. The system binds a unique tag system to each layered nursing operation unit, using the corresponding operation time period as the time sequence tag and the corresponding operation configuration as the attribute tag, so that each layered nursing operation has exclusive tag characteristics that are identifiable, locatable, and independently retrievable.

[0036] After specific stratification, an independent replacement mechanism for each stratification is established. Based on real-time dynamic monitoring data of patients and real-time risk changes, different stratified nursing operation units are independently judged, matched, and replaced and updated. The specific process is to detect the marked correction data, identify the labels corresponding to the correction data, match the time stratification corresponding to the operation time period in the label, match the standard nursing configuration of the corresponding time stratification, and compare it with the stratified nursing operation to obtain the configuration correction data that needs to be corrected.

[0037] Based on the configured correction data, the system locates the preset nursing collaborative operation plan corresponding to the target stratum and repeatedly performs dynamic optimization of the preset collaborative operation. It determines whether operation replacement is needed. If an operation replacement instruction is triggered, the system calls the preset backup nursing operation template of the same stratum in the stratum library to accurately replace the target stratum nursing unit, and completes the update of the corresponding stratum operation content, modification of monitoring frequency, adjustment of intervention intensity, and refresh of multi-role collaborative tasks. After the replacement is completed, the system solidifies the new stratum operation content in real time and updates the execution sequence and quality control standards of the stratum unit simultaneously, so that the stratum nursing operation corresponding to the current time period and current risk level is fully adapted to the patient's real-time condition.

[0038] Some of the data in the above formula are calculated by removing dimensions and taking their numerical values. The formula is the closest to the real situation obtained by software simulation of a large amount of collected data. The preset parameters and preset thresholds in the formula are set by those skilled in the art according to the actual situation or obtained through simulation of a large amount of data.

[0039] Working principle of the invention: The system first simultaneously collects intraoperative static data, dynamic baseline data, and specialized nursing plan data matching the surgical type for patients undergoing cardiovascular interventional surgery, and continuously collects real-time dynamic monitoring data of patients after surgery. By quantifying and assigning values ​​to the three types of core data, static risk scores, dynamic correction scores, and nursing risk scores are obtained. The static risk scores and nursing risk scores are then normalized, coupled, and fused with hierarchical weights to obtain a static risk estimate. The baseline is then denoised and corrected by combining the deviation disturbance of the dynamic correction score, outputting an accurate postoperative risk estimate for the patient. The system classifies postoperative risk levels based on the combined risk estimate, presets standardized PACU nursing collaborative operations adapted to different risk levels as needed, and performs temporal arrangement and adaptive logic embedding of the preset operations. At the same time, relying on the trained neural network operation optimization model, the system verifies the adaptability of the current nursing plan to the patient's dynamic condition in real time. Based on the patient's real-time fluctuations in vital signs and changes in risk, the system dynamically optimizes and adjusts the preset nursing collaborative operations and records and retains the entire process.

[0040] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. An intelligent collaborative nursing system for cardiovascular interventional patients in PACU, characterized in that, include: Acquire surgical data during cardiovascular interventional procedures and nursing data after surgery, and collect dynamic monitoring data of patients in real time; By combining surgical and nursing data, a risk assessment is conducted on the patient's condition after cardiovascular interventional surgery, and based on the results of the risk assessment, collaborative nursing procedures that the patient needs are planned. The preset collaborative operations are dynamically optimized and adjusted based on dynamic monitoring data, and the results are recorded.

2. The intelligent collaborative nursing system for cardiovascular interventional patients in PACU according to claim 1, characterized in that, The risk assessment of patients after cardiovascular interventional surgery, combining surgical and nursing data, includes: Extract surgical and nursing data; the surgical data includes intraoperative static data and dynamic baseline data; the nursing data consists of nursing plans developed based on the type of surgery. Intraoperative static data are quantified to obtain static risk scores, dynamic baseline data are quantified to obtain dynamic correction scores, and nursing data are quantified to obtain nursing risk scores. By combining the static risk score and the nursing risk score, the static risk estimate of the patient's surgery is obtained. The dynamic correction score is used as the impact data of the static risk estimate to correct it, resulting in the combined risk estimate.

3. The intelligent collaborative nursing system for cardiovascular interventional patients in PACU according to claim 2, characterized in that, The method of combining and superimposing static risk scores with nursing risk scores to obtain a static risk assessment of a patient's surgery includes: The static risk score and the nursing risk score are normalized, and the normalized static risk score and the nursing risk score are coupled and merged to obtain the co-coupled risk value. The risk prediction level is obtained by comparing the collaborative coupling risk value with the preset risk threshold. Based on the risk prediction level, corresponding level weight coefficients are set for the static risk score and the nursing risk score. The static risk estimate is obtained by combining the synergistic coupling risk value, static risk score, and nursing risk score with their respective grade weight coefficients.

4. The intelligent collaborative nursing system for cardiovascular interventional patients in PACU according to claim 2, characterized in that, The process of adjusting the dynamic correction score as impact data for static risk valuation to obtain a combined risk valuation includes: The dynamic correction score is normalized, and the normalized dynamic correction score is compared with the standard dynamic correction value in the database to obtain the deviation disturbance. Using static risk valuation as a fixed risk baseline, the combined risk valuation is obtained by superimposing and correcting the fixed risk baseline through deviation disturbance.

5. The intelligent collaborative nursing system for cardiovascular interventional patients in PACU according to claim 1, characterized in that, The pre-setting of collaborative procedures for patient care based on risk assessment results includes: The postoperative risk level of cardiovascular interventional surgery will be determined by combining risk assessment. For different postoperative risk levels, corresponding standardized and specialized PACU nursing basic operations and key special operations will be preset to form a graded nursing operation library. All preset nursing collaborative operations in the hierarchical nursing operation library are arranged in a time-sequential and process-oriented manner, and adaptive replacement logic is added to the steps of all preset nursing collaborative operations. When marked correction data is detected, the preset nursing collaborative operations are adaptively adjusted. Among them, the correction data is the data when the surgical data and nursing data are updated, and the marked correction data is the data that has a significant impact on the preset nursing collaborative operations.

6. The intelligent collaborative nursing system for cardiovascular interventional patients in PACU according to claim 1, characterized in that, The process of dynamically optimizing and adjusting preset collaborative operations based on dynamic monitoring data and recording the results includes: The operation optimization model is obtained by training a neural network model using standard monitoring data, preset collaborative operations, and corresponding adjustment operation data from the database as training data. By incorporating real-time dynamic monitoring data into the operation optimization model, the model updates patient risk assessment and risk level in real time, and simultaneously verifies the adaptability of the currently implemented preset nursing collaborative operation plan, and performs hierarchical adaptive optimization and adjustment of the preset nursing collaborative operation.

7. The intelligent collaborative nursing system for cardiovascular interventional patients in PACU according to claim 6, characterized in that, The process of training a neural network model using standard monitoring data from a database, preset collaborative operations, and corresponding adjustment operation data to obtain an operation optimization model includes: Select a model framework and deep learning algorithm from the artificial intelligence library; build a model based on the model framework and deep learning algorithm to obtain the initial model; Obtain the standard dataset; the standard dataset includes standard input data consistent with the content attributes of the state analysis sequence, and standard output data consistent with the content attributes of the state labels; The standard dataset is divided into a training set, a validation set, and a test set according to a set ratio; the initial model is trained using the training set; the internal structure and parameters of the initial model are adjusted using the validation set; and the trained initial model is tested using the test set to obtain test metrics. Obtain the indicator threshold; compare the test indicator with the indicator threshold; if all test indicators are greater than the indicator threshold, mark the initial model as the operation optimization model; otherwise, rebuild and retrain the initial model.

8. The intelligent collaborative nursing system for cardiovascular interventional patients in PACU according to claim 6, characterized in that, The synchronization involves performing an adaptability check on the currently executing preset nursing collaborative operation plan, including: The system incorporates real-time dynamic monitoring data into the risk assessment model, updates the patient's combined risk estimate and risk level in real time, integrates the combined risk estimate and risk level to obtain the risk status, and performs an adaptability check on the currently executed preset nursing collaborative operation plan based on the updated risk status. It determines from multiple dimensions whether the current nursing plan matches the patient's real-time condition. When there is an indicator deviation or risk level rise or fall, a dynamic optimization and adjustment mechanism is triggered.

9. The intelligent collaborative nursing system for cardiovascular interventional patients in PACU according to claim 6, characterized in that, The hierarchical adaptive optimization adjustment of the preset nursing collaborative operations includes: Pre-defined collaborative nursing operations are stratified based on the principles of operation time period and operation configuration to obtain stratified nursing operations. The operation time period and operation configuration are used as tags and corresponding stratified nursing operations. The system detects the labeled correction data, matches the configuration correction data according to the labels of the labeled correction data, matches and compares the configuration correction data with the preset nursing collaborative operation to generate operation replacement commands, identifies the operation replacement commands, and performs hierarchical operation replacement optimization adjustments.

10. The intelligent collaborative nursing system for cardiovascular interventional patients in PACU according to claim 9, characterized in that, The step of matching and comparing the configuration correction data with preset nursing collaborative operations to generate operation replacement commands includes: Identify the labels corresponding to the corrected data, match them with the corresponding standard nursing configurations, and compare them with the stratified nursing operations to obtain the configuration correction data that needs to be corrected; Match the nursing collaboration operations corresponding to the target stratum of the configuration correction data, and compare and verify the compatibility to determine whether a replacement operation is needed; if yes, generate an operation replacement instruction; otherwise, identify the next time period within the stratum.

Images