A cloud computing-based cardiology nursing information monitoring system

Abstract

The application relates to the technical field of informationization of cardiology nursing, and discloses a cardiology nursing information monitoring system based on cloud computing. A sign monitoring module of the system integrates real-time data streams of a wearable device and a bedside monitor, collects multi-dimensional physiological parameters, and generates a dynamic vital sign set; a health assessment module identifies abnormal fluctuation nodes of sign data, matches historical medical record, quantifies the correlation strength between current signs and typical pathological patterns, and generates multi-dimensional health risk indicators; a resource allocation module analyzes the nursing resource occupation state based on the risk indicators, calculates the task urgency weight, dynamically allocates the work path of medical staff and the equipment use sequence, and generates a hierarchical nursing scheduling scheme; and an intervention analysis module executes the scheduling scheme, compares the sign changes before and after the resource allocation, detects the abnormal response delay time, locates the execution deviation nodes, and generates a nursing effect traceability report. The system helps the intelligentization and fine management of the cardiology nursing process.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of information technology for cardiology nursing, in particular to a cardiology nursing information monitoring system based on cloud computing. BACKGROUND

[0002] In the clinical nursing of cardiology, the monitoring of physiological parameters of patients, the assessment of health status, the allocation of nursing resources and the tracing of intervention effects are always the key links to ensure the quality of nursing. Under the traditional nursing mode, the collection of patient's physical data mainly depends on manual recording or single device monitoring, and the data is scattered in different recording carriers or device systems, which is difficult to realize centralized integration and real-time tracking. For example, the heart rate data recorded by wearable devices, the blood pressure fluctuation curve and the change of blood oxygen saturation monitored by bedside monitors are often independent of each other, and medical staff need to switch between different platforms to query, which not only increases the workload of data integration, but also may cause the judgment of patient's condition to lag due to data transmission delay.

[0003] For the assessment of patient's health status, the traditional method mainly depends on the personal experience and subjective judgment of medical staff, and lacks systematic mining and quantitative analysis of historical medical record data. When the patient's physical signs are abnormal, the medical staff need to manually check the records of past health events, which is difficult to quickly match the typical pathological pattern, resulting in the lack of unified standard for the assessment of the deviation degree of health status, and the easy occurrence of missed or misjudged cases. Especially in the scene of complex and changeable patient's condition in cardiology, this experience-dependent assessment method is difficult to meet the needs of precise nursing.

[0004] In terms of nursing resource allocation, the traditional mode usually adopts fixed scheduling or allocation according to bed number, without fully considering the emergency degree of patient's condition and the real-time occupation state of resources. When multiple patients have nursing needs at the same time, the arrangement of medical staff's work path and medical equipment use sequence lacks dynamic adjustment mechanism, which may cause high-risk patients to be unable to get priority intervention, and low-emergency tasks to occupy too many resources, resulting in waste of nursing resources and low efficiency of scheduling.

[0005] After the implementation of nursing intervention measures, the tracing and analysis of its effect also have deficiencies. Under the traditional mode, the comparison of physical signs before and after intervention, the recording of abnormal response delay time and the positioning of execution deviation node mainly depend on manual summary, and the data integrity and accuracy are difficult to guarantee, which leads to the inability to find problems in nursing process in time and the difficulty to realize the continuous optimization of nursing quality. The existence of these problems makes the intelligentization and refinement level of cardiology nursing work need to be improved, and an integrated information monitoring system integrating data collection, evaluation, scheduling and analysis is urgently needed. SUMMARY

[0006] The purpose of this invention is to provide a cloud computing-based cardiology nursing information monitoring system to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides a cloud computing-based cardiology nursing information monitoring system, the system comprising:

[0008] The vital signs monitoring module collects multidimensional physiological parameters of cardiology patients, integrates real-time data streams from wearable devices and bedside monitors, continuously tracks the patient's heart rate variability and blood pressure fluctuation curves, calculates abnormal deviations in respiratory rate, and generates a dynamic set of vital signs by combining changes in blood oxygen saturation.

[0009] The health assessment module receives the dynamic set of vital signs, identifies abnormal fluctuation nodes in the vital sign data, matches health event records in the patient's historical medical records, quantifies the correlation strength between the current vital signs and typical pathological patterns, assesses the degree of deviation of the patient's health status, and generates multi-dimensional health risk indicators.

[0010] Based on the multi-dimensional health risk indicators, the resource allocation module analyzes the real-time occupancy status of nursing resources, calculates the urgency weight of different nursing tasks, dynamically allocates the work paths of medical staff and the usage sequence of medical equipment, and generates a hierarchical nursing scheduling scheme.

[0011] The intervention analysis module executes the hierarchical nursing scheduling plan, compares the changes in vital sign data before and after resource allocation, detects the response delay time of abnormal vital signs, locates the deviation nodes in the execution of nursing intervention measures, and generates a nursing effect traceability report.

[0012] Preferably, the vital sign monitoring module includes:

[0013] The multi-source fusion submodule synchronously receives heterogeneous data streams from an electrocardiogram monitor, a pulse oximeter, and a non-invasive blood pressure monitor, calibrates the sampling timestamps of different devices, extracts the extreme points of physiological parameters within five consecutive minutes, calculates the synchronization index of parameter fluctuation frequency, and generates a time-aligned set of vital signs data.

[0014] The anomaly detection submodule analyzes the time-aligned vital sign data set, divides the baseline parameter range of the day-night cycle, monitors the duration of real-time parameters exceeding the range threshold, counts the occurrence density of abnormal events per unit time, and generates a vital sign anomaly probability matrix.

[0015] The risk integration submodule calls the abnormal vital signs probability matrix, associates the drug metabolism time window in the patient's medication record, calculates the correlation coefficient between abnormal vital signs and drug concentration, integrates external interference factors from environmental temperature and humidity sensors, and generates the dynamic vital signs set.

[0016] Preferably, the health assessment module includes:

[0017] The pathological matching sub-module extracts the heart rate oscillation parameters in the dynamic vital sign set, compares the waveform feature library of typical heart failure cases, calculates the spectral difference degree of the current waveform and the pathological reference waveform, and identifies the occurrence position and duration interval of the premature beat event;

[0018] The risk quantification sub-module receives the spectral difference degree and premature beat event data, loads the historical values of myocardial enzyme spectrum in the patient's electronic medical record, constructs the dynamic change curve of the current myocardial injury indicator, and quantifies the proximity degree of the curve slope and the critical value;

[0019] The indicator generation sub-module integrates the dynamic change curve of the myocardial injury indicator and the pathological matching result, weightedly calculates the priority order of the risk of organ failure, and generates the multi-dimensional health risk indicator including the heart function classification label and the organ risk weight.

[0020] Preferably, the resource allocation module comprises:

[0021] The task analysis sub-module decomposes the heart function classification label in the multi-dimensional health risk indicator, maps the nursing operation list corresponding to different levels, identifies the inventory status of emergency drugs and the occupation situation of respirators, and calculates the device dependency coefficient of each nursing operation;

[0022] The path optimization sub-module receives the nursing operation list and device dependency coefficient, constructs a real-time topological graph of medical staff location information, simulates the task completion time delay rate under different paths, and iteratively optimizes the execution order of the nursing operation sequence;

[0023] The scheduling generation sub-module integrates the optimized nursing operation sequence and device allocation scheme, coordinates the use time conflict of intravenous infusion pumps and defibrillators, and generates the hierarchical nursing scheduling scheme including the operation time window and the device handover node.

[0024] Preferably, the intervention analysis module comprises:

[0025] The response monitoring sub-module captures the real-time sign data stream after the hierarchical nursing scheduling scheme is executed, marks the corresponding relationship between the diuretic administration time point and the blood pressure drop event, and counts the delay duration of the sign parameter returning to the safety threshold;

[0026] The deviation positioning sub-module compares the time stamps of the nursing operation preset standard and the execution record, identifies the operation node where the intravenous injection speed deviates from the standard value, and associates the deviated operation with the blood potassium concentration mutation event in the adjacent time window;

[0027] The traceability generation sub-module integrates the delay duration and the occurrence position of the blood potassium concentration mutation event, constructs a causal chain model of nursing measures and sign response, and generates the nursing effect traceability report including the key intervention node execution deviation record.

[0028] Preferably, the system further comprises:

[0029] The early warning execution module receives the key intervention node execution deviation record in the nursing effect traceability report, calls the operation history database of the corresponding medical staff, matches the validity period state of the emergency skill certification certificate, and generates a personalized skill reinforcement training scheme;

[0030] The personalized skill reinforcement training scheme includes an operation procedure simulating an emergency scene, a device operation standard video tutorial, and an examination and evaluation index.

[0031] Preferably, the early warning execution module comprises:

[0032] The ability evaluation submodule analyzes the operation type of the key intervention node execution deviation record, retrieves the training completion rate data in the medical staff archive, and calculates the correlation strength between operation errors and skill certificate update time;

[0033] The training construction submodule configures a virtual reality simulation cardiogenic shock scene based on the operation type correlation strength result, and dynamically generates a correction training module for infusion rate adjustment error;

[0034] The scheme generation submodule integrates the virtual reality simulation scene and the correction training module, superimposes the operation weak point analysis report in the historical examination, and generates the personalized skill reinforcement training scheme including training class hour allocation and examination threshold.

[0035] Preferably, the vital sign monitoring module further comprises:

[0036] The environmental calibration submodule obtains real-time readings of the ward temperature and humidity sensor, calculates the influence coefficient of temperature fluctuation on vasoconstrictor, and corrects the environmental interference deviation amount of blood pressure monitoring value;

[0037] The drug interference analysis submodule associates the vasoactive drug infusion rate in the electronic medical order system, quantifies the lag response time of drug concentration peak and heart rate variation, and generates a drug influence compensation parameter;

[0038] The data correction submodule fuses the environmental interference deviation amount and the drug influence compensation parameter, reconstructs the reference value of the original vital sign data, and updates the blood pressure fluctuation curve in the dynamic vital sign set.

[0039] Preferably, the path optimization submodule performs operations including:

[0040] The real-time position tracking unit obtains indoor positioning beacon data worn by medical staff, and constructs a three-dimensional space coordinate mapping model of the ward;

[0041] The task conflict detection unit identifies the geographical position radius of multiple emergency tasks in the same period, calculates the resource competition probability of the path intersection area;

[0042] The dynamic adjustment unit reassigns the execution subject of the nursing task based on the resource competition probability result, to avoid overlapping conflicts of equipment transportation paths.

[0043] Preferably, the system further comprises:

[0044] The medication safety monitoring module receives the blood potassium concentration data in the dynamic vital sign set, associates the diuretic infusion rate parameter being executed by the electronic medical order system, identifies the potential risk time window of the drug contraindication combination, and generates a high-risk drug compatibility warning instruction;

[0045] The high-risk drug compatibility warning instruction triggers the automatic locking mechanism of the infusion pump and synchronously pushes the alternative treatment scheme to the mobile nursing terminal.

[0046] Compared with the prior art, the beneficial effects of the present application are:

[0047] The cardiology nursing information monitoring system based on cloud computing brings multiple optimizations to cardiology nursing work through the cooperative operation of multiple modules. In the aspect of vital sign data collection, the vital sign monitoring module breaks through the limitations of scattered data of wearable devices and bedside monitors in traditional nursing, integrates real-time data streams, continuously tracks key physiological parameters such as heart rate variability and blood pressure fluctuation curve, and generates a dynamic vital sign set. This process realizes the centralized management of multi-source data, enables medical staff to obtain comprehensive physiological state information of patients on a unified platform, avoids the tediousness of querying one by one in different devices or records, reduces the possibility of data omission or transmission delay, and makes the perception of patient condition changes more timely.

[0048] The health assessment module identifies abnormal fluctuation nodes in the vital sign data, combines health event records in the patient's historical medical record, quantifies the association strength between the current vital sign and the typical pathological pattern, and generates multi-dimensional health risk indicators. This assessment method no longer simply relies on the personal experience of medical staff, but relies on data association analysis to form objective risk quantification results. Medical staff can more clearly judge the deviation degree of the patient's health status according to these indicators, which provides a more reliable reference for formulating targeted nursing schemes, helps to reduce the evaluation deviation caused by subjective judgment differences, and improves the consistency and accuracy of health assessment.

[0049] The resource allocation module operates based on multi-dimensional health risk indicators. By analyzing the real-time occupancy status of nursing resources and calculating the urgency weight of different nursing tasks, it dynamically allocates work paths for medical staff and usage sequences for medical equipment. This dynamic scheduling mechanism changes the traditional fixed-schedule or bed-based resource allocation model, enabling resource allocation to closely align with the actual urgency of patients' needs. Nursing tasks for high-risk patients receive priority resource support, medical staff work path planning becomes more rational, and the utilization efficiency of medical equipment is improved. This avoids resource idleness or over-concentration on low-urgency tasks, allowing limited nursing resources to be used more effectively.

[0050] During the execution of the tiered nursing care scheduling plan, the intervention analysis module compares changes in vital sign data before and after resource allocation, detects response delays to abnormal vital signs, and identifies deviations in the implementation of nursing interventions. This step provides data support for the traceability of nursing outcomes. By analyzing various nodes in the execution process, medical staff can identify potential problems in the nursing workflow, such as untimely interventions or improper equipment use. Based on these findings, targeted adjustments and optimizations can be made to the nursing workflow, continuously improving nursing operation standards and promoting continuous improvement in nursing quality, thus forming a closed loop of "monitoring-assessment-intervention-analysis-optimization" throughout the entire nursing process. Attached Figure Description

[0051] Figure 1 This is a schematic diagram illustrating the working principle of the cloud computing-based cardiology nursing information monitoring system described in this invention.

[0052] Figure 2 This is a schematic diagram illustrating the working principle of the health assessment module submodule.

[0053] Figure 3 A schematic diagram illustrating the working principle of the path optimization submodule;

[0054] Figure 4 This is a schematic diagram illustrating the working principle of the early warning execution module. Detailed Implementation

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] Please see Figure 1 This invention provides a cloud computing-based cardiology nursing information monitoring system, the system comprising:

[0057] The cloud computing platform integrates real-time physiological monitoring data of patients in the cardiology department with nursing resource scheduling functions to build a closed-loop intelligent nursing management system. The physical sign monitoring module uses multi-device collaborative collection technology to continuously obtain patient heart rate variability, blood pressure fluctuation curve, respiratory rate offset, and blood oxygen saturation, and other parameters to form a dynamically updated set of vital signs. The health assessment module uses pathological pattern matching algorithms to correlate and analyze real-time vital sign data with historical health events in medical records to generate multi-dimensional assessment indicators including cardiac function classification and organ risk weight. The resource allocation module dynamically optimizes the path of medical staff and the sequence of device use based on the priority of risk indicators and the real-time occupancy status of hospital nursing resources. The intervention analysis module uses timestamp comparison and causal chain modeling to quantify and trace the effectiveness of nursing measures. The modules interact through a cloud service bus and use a microservices architecture to ensure system scalability.

[0058] Example 1: see Figure 2 The physical sign monitoring module uses multi-source data fusion technology to achieve real-time collection and integration of physiological parameters of patients in the cardiology department. The module interfaces with various monitoring devices in the hospital, including electrocardiogram monitors, pulse oximeters, non-invasive blood pressure meters, and other medical instruments with different sampling frequencies and communication protocols. The data acquisition unit uses an asynchronous communication mechanism to receive raw signals sent by each device through standardized data interfaces, including electrocardiogram waveforms, blood oxygen saturation percentage, and arterial blood pressure values. To address the clock bias problem of different devices, the system introduces a reference time axis in the data preprocessing stage and uses interpolation algorithms to align the time sequence of discrete sampling points, ensuring that all parameters have a unified time reference. Within a five-minute data window, the system automatically identifies the extreme points of each parameter, including maximum heart rate, minimum blood oxygen, and blood pressure fluctuation peak, and calculates the synchronicity indicators of these extreme values to assess the coordination of multi-system physiological responses.

[0059] The abnormality detection function is achieved by establishing a dynamic threshold model. The system sets the reference parameter range for each patient based on individual differences and historical data, including the safe range of heart rate, the reasonable fluctuation amplitude of blood pressure, and the minimum threshold of blood oxygen saturation. During real-time monitoring, the system continuously tracks the duration of each parameter deviating from the reference range, and when a certain indicator exceeds the threshold and maintains a preset duration, an abnormal event counter is triggered. This counter not only records the frequency of abnormal events, but also calculates the density of abnormal events per unit time to form a probability matrix reflecting the stability of the patient's state. To distinguish the clinical significance of abnormal events, the system analyzes the correlation between drug metabolism period and physiological parameter fluctuations based on the patient's medication record to distinguish pathological changes from normal responses caused by drug intervention.

[0060] The risk integration function further introduces environmental factors. The system accesses the temperature and humidity sensor network in the ward to obtain real-time data of the microenvironment where the patient is located. Considering that blood pressure measurements are susceptible to temperature, the system establishes an environmental compensation algorithm to dynamically correct the original blood pressure data based on the difference between the current temperature and the standard value. At the same time, for patients receiving treatment with vasoactive drugs, the system reads the drug infusion rate from the electronic medical order and, combined with the pharmacokinetic characteristics of the drug, calculates the expected impact of drug concentration on heart rate variability and generates the corresponding compensation parameters. Finally, the system integrates the physiological parameters that have been environmentally corrected and drug compensated into a structured vital sign set, which not only contains the original monitoring values but also includes multi-dimensional labels such as data source, environmental conditions, and drug effects, providing a comprehensive and reliable data foundation for subsequent health assessment.

[0061] The health assessment module achieves accurate judgment of patient status through multi-level pathological feature analysis. The pathological matching function first performs multi-scale decomposition on the electrocardiogram signal to extract the morphological features and time-domain characteristics of the waveform. The system maintains a cloud-based pathological feature library containing typical waveform patterns of various heart diseases. Through the dynamic time warping algorithm, the system calculates the similarity between the patient's real-time electrocardiogram signal and the pathological patterns in the feature library to identify specific segments of abnormal waveforms and their duration. For the identified abnormal waveforms, the system further analyzes their position density and time distribution to determine whether they constitute clinically significant premature events or rhythm disorders.

[0062] The risk quantification function focuses on the dynamic assessment of myocardial injury. The system regularly extracts the patient's myocardial enzyme spectrum historical data, including troponin and creatine kinase isoenzyme, from the electronic medical record. Through sliding window analysis technology, the system tracks the trend of these indicators over time and calculates the rate of change within a specific time interval. When the indicator value shows a consistent upward or downward trend and the rate of change exceeds the pre-set warning threshold, the system activates the corresponding risk warning mechanism. This trend-based warning method can detect potential risk signals in advance before the absolute value reaches the critical standard.

[0063] The index generation function adopts a multi-parameter fusion strategy. The system receives analysis results from the pathological matching and risk quantification sub-modules, considering factors such as heart rate variability characteristics, premature event density, and myocardial injury trend. Through pre-set weight distribution rules, the system calculates the risk score of each aspect of cardiac function and provides correction suggestions for cardiac function classification based on the score results. At the same time, the system assesses the potential secondary effects of cardiac function abnormalities on other organs, especially the potential risks to kidney perfusion and function, and generates a comprehensive risk assessment report reflecting the interaction between multiple organ systems. The report uses a structured data format, including machine-readable risk level labels and detailed assessment basis, providing objective support for clinical decision-making.

[0064] Throughout the implementation process, the system adopts a distributed computing architecture to handle massive physiological data and complex analysis algorithms. The cloud computing platform provides elastic computing resources, which can dynamically adjust processing capacity according to real-time data volume. All analysis results are verified by the prior knowledge of clinical experts to ensure that the algorithm output conforms to medical logic. The system also establishes a perfect data quality control mechanism, including signal quality detection, outlier filtering, missing data compensation and other technical means, to ensure the reliability of input data and the accuracy of analysis results. Through this comprehensive and meticulous implementation method, the system realizes real-time monitoring and accurate assessment of the health status of patients in the cardiology department.

[0065] Example 2: refer to Figure 3 , the resource allocation module realizes the efficient scheduling and management of nursing resources in the cardiology department through intelligent task analysis and path optimization algorithms. This module receives structured risk indicators from the health assessment module and converts them into executable nursing task sequences. The system's built-in cardiac function grading mapping table establishes a correspondence between different levels of risk assessment results and standard nursing operations, such as automatically triggering 12 basic nursing measures for patients with cardiac function IV level, including non-invasive ventilation support, intravenous diuretic injection, continuous electrocardiogram monitoring, etc. During task analysis, the system queries the RFID tag data of the hospital material management system in real time to obtain critical resource information such as emergency drug inventory status, ventilator usage, etc., and calculates the device dependency and material consumption of each nursing operation.

[0066] After task decomposition is completed, the path optimization function begins to operate. The system tracks the location information of medical staff in real time through the UWB ultra-wideband positioning network deployed in the ward. The positioning beacon worn by each nursing staff updates coordinate data at a frequency of 1Hz, and the system maps these spatial information into a three-dimensional ward model to construct a dynamically updated location topology graph. When multiple emergency tasks occur simultaneously, the system calculates the geographical radius between the execution locations of each task and analyzes the degree of overlap of the time window. If the distance between the geographical centers of two emergency tasks is less than the preset threshold and the time window overlap exceeds a certain length of time, the system determines that there is a path conflict risk. At this time, the dynamic adjustment algorithm begins to re-evaluate the task allocation scheme, considering factors such as the professional skill level of nursing staff, the distance between the current location and the task point, the straight-line accessibility of equipment transportation path, etc., to generate the optimal task redistribution scheme.

[0067] The scheduling generation function is responsible for converting the optimized task sequence into an executable care plan. The system establishes a device usage time conflict matrix to analyze the occupancy requirements of different care operations on medical devices. For parallel tasks that require the same device resources, the system automatically inserts buffer time to ensure the orderly handover of critical devices such as infusion pumps and electrocardiogram monitors. When generating the final scheduling plan, the system not only considers time factors but also assesses the clinical priority between tasks to ensure that the care needs of high-risk patients are prioritized. The scheduling plan is displayed in a visual form on the mobile care terminal, including detailed execution time windows, device handover nodes, operation precautions, and other information to facilitate accurate understanding and execution by nursing staff.

[0068] The application of indoor positioning technology provides accurate spatial data support for path optimization. The UWB positioning base stations deployed by the system cover the entire cardiology department, and the distance between the tag and the base station is calculated by measuring the time of flight of the wireless signal. The three-dimensional spatial mapping model converts the physical ward into a digital coordinate system, accurately restoring the spatial relationship of key areas such as patient rooms, nurse stations, and equipment rooms. Positioning data is processed through Kalman filtering to eliminate interference factors such as multipath effects, ensuring the stability and reliability of location information. The system regularly calibrates positioning accuracy by comparing the coordinates of reference points to adjust positioning parameters and maintain centimeter-level positioning accuracy.

[0069] Task conflict detection adopts a space-time dual judgment standard. The system defines the influence range for each emergency task, with the execution location as the center to determine the service radius. When the service areas of different tasks intersect in space and there is overlap in planned execution time, the system starts the conflict resolution process. Conflict probability calculation not only considers geographic distance but also includes task duration, resource demand intensity, and other variables to form a multi-dimensional conflict assessment model. The system maintains a dynamically updated task queue to monitor the execution progress of each task in real time and adjusts the sensitivity threshold of conflict judgment based on actual conditions.

[0070] Dynamic adjustment function is based on combination optimization algorithm to realize resource reallocation. The system uses an improved Hungarian algorithm to handle the task-person matching problem, considering not only path distance but also professional skill matching degree, current work load and other constraint conditions. For device transportation path planning, the system calculates the length, number of turns, and difficulty of passing of each alternative path to select the most comprehensive optimal transportation scheme. When detecting the transportation path of large equipment such as ventilators crossing, the system automatically adjusts the transportation time or selects a detour route to avoid corridor congestion. The adjusted scheme is pushed to the mobile terminal of relevant nursing staff in real time, displaying the updated task list and navigation path.

[0071] Device scheduling management ensures rational use of resources through timeline conflict detection. The system establishes a usage timeline for each critical medical device, marking the occupied time period and the expected release time. When a new care task applies for the use of a device, the system checks the overlap between the application time period and the existing occupation. For complete conflict, the system automatically coordinates the use order according to the task priority; for partial conflict, the system tries to fine-tune the operation time or suggests the use of alternative devices. The management of commonly used devices such as infusion pumps adopts a reservation system, and nursing staff can check the device availability through mobile terminals and make advance reservations. The system automatically records the device handover process, including handover time, device status confirmation, and transfer of key information such as usage precautions, forming a complete device flow log.

[0072] The interaction design of the mobile terminal fully considers the needs of the clinical work scene. The tablet computer equipped by the nursing staff is installed with a special application program, which displays the personal task list in the form of a timeline. The task card displays the patient's basic information, nursing measures content, required devices, and estimated time consumption. The navigation function provides optimal path guidance from the current location to the task point, and updates the obstacles or congestion on the path in real time. During the task execution process, the system supports voice interaction and gesture operation, which is convenient for nursing staff to operate the system in a sterile environment. After completing the task, the nursing staff confirms the execution by scanning the patient's wristband and drug bar code, and the system automatically updates the task status and releases the related resources.

[0073] The entire resource allocation process adopts a closed-loop control mechanism. The system continuously monitors the progress of care tasks, compares the actual completion time with the planned time, and detects significant delays. When a significant delay is detected, the system analyzes the delay reasons and triggers the corresponding compensation mechanism, such as increasing personnel, adjusting the time window of subsequent tasks, etc. The execution details of completed tasks, including operation time, execution personnel, and used devices, are recorded completely, which are used for subsequent quality analysis and algorithm optimization. The system regularly evaluates the actual execution effect of the scheduling scheme, continuously adjusts and optimizes the parameters through the analysis of historical data, and improves the accuracy and practicality of resource allocation.

[0074] With this refined implementation, the resource allocation module realizes the intelligent management of the entire process of cardiology nursing work from task decomposition, path planning to execution monitoring. The system converts abstract nursing needs into specific executable plans, coordinates the rational distribution of human and equipment resources in time and space, and effectively improves the efficiency and quality of nursing work. The combination of three-dimensional space modeling and real-time positioning technology provides accurate spatial data support for resource scheduling in complex medical environments. The dynamic adjustment algorithm can quickly respond to emergencies and ensure the rational allocation of emergency resources. The visual interactive interface reduces the technical threshold of system use, allowing nursing staff to focus on clinical operations rather than system operations. The entire module runs independently of any pre-set fixed rules, but through continuous learning and adaptation, it forms a dynamic scheduling strategy that matches the specific work environment.

[0075] Example 3: see Figure 4 The intervention analysis module realizes the quantitative evaluation and tracing of the effect of clinical intervention by establishing a correlation model between nursing measures and patient physiological responses. This module is started after the implementation of the hierarchical nursing scheduling scheme and continuously monitors the dynamic vital signs of patients, capturing the timing relationship between key intervention nodes and physiological parameter fluctuations. The system uses a high-frequency data acquisition mode, shortening the physiological parameter sampling interval from the regular 5 minutes to 30 seconds after key operations such as intravenous drug administration and body position adjustment, enhancing the ability to capture rapid changes. The diuretic administration event triggers a special blood pressure monitoring protocol, and the system records the administration timestamp and establishes a two-hour data observation window before and after the event, analyzing the change trajectory of systolic, diastolic, and mean arterial pressures.

[0076] Physiological response analysis uses a sliding window comparison method, and the system calculates the statistical feature difference of blood pressure parameters before and after the intervention. When the systolic pressure is detected to have decreased by more than a certain percentage of the baseline value within an hour, the system marks this event as a positive drug response case. Time series alignment algorithm accurately matches the nursing operation log and physiological data stream, identifying deviations in intravenous infusion speed, drug concentration, and other execution parameters from standard operating procedures. The push rate difference calculation formula is as follows:

[0077]

[0078] where, represents the actual recorded push completion time, represents the push time required by the standard operating procedure. When the calculation result exceeds the pre-set threshold, the system marks this deviation event on the time axis and extracts the blood potassium detection results within a specific time window before and after the event for correlation analysis.

[0079] The deviation traceability function is realized by constructing a multi-factor influence network. The system integrates multiple sources of data such as nursing operation deviations, environmental parameter fluctuations, and drug interactions to establish a Bayesian probability graph model. This model takes intravenous bolus rate deviation, room temperature changes, and concurrent medication as parent nodes, and blood potassium concentration mutation as a child node, to calculate the conditional probability of each factor on abnormal physiological response. Network parameters are trained through historical case data, continuously optimizing the weight relationship between nodes. When a new abnormal event occurs, the system runs a probability reasoning algorithm to generate a causal analysis report containing the contribution of each factor. The report displays key influencing factors in the form of a heat map and sorts them according to clinical relevance, helping medical staff quickly locate potential problem sources.

[0080] The warning execution module initiates targeted improvement measures based on the traceability analysis results. The system retrieves the electronic archives of nursing staff related to operation deviations, analyzes their training records and skill certification status. When it finds that the operator has not completed the annual simulation examination or specific skill training, the system automatically generates a reinforcement training program containing the missing course content. The training program uses adaptive logic to dynamically adjust training content and assessment standards based on deviation type and severity. For intravenous bolus operation deviations, the program focuses on configuring injection pump calibration, drug calculation, and other practical modules; for emergency response delays, it increases the frequency and difficulty of scenario simulation drills.

[0081] The construction of the virtual training environment adopts a modular design principle. The system develops a configurable virtual scene based on the Unity3D engine, simulating common emergency conditions in the cardiology department. The vascular active drug compounding training module reproduces blood pressure response curves under different dosing regimens, and students need to adjust infusion parameters based on real-time changes in vital signs. The fluid resuscitation training scene includes dynamic disease evolution logic, and each operation decision made by the student will affect the virtual patient's outcome path. The system records operation details during training, including decision-making time, operation accuracy, and process compliance, and generates a personalized evaluation report containing improvement suggestions.

[0082] Skill assessment uses a multi-dimensional quantitative index system. The system not only assesses the final result of the operation, but also analyzes the performance of key nodes in the operation process. The assessment of intravenous drug administration includes drug calculation accuracy, device parameter setting rationality, and execution time control precision. Each indicator sets a dynamic threshold, adjusting the standard for meeting the requirements according to the student's job level and experience level. Evaluation data is connected to the hospital continuing education credit system, and nursing staff who complete training can obtain corresponding professional development credits.

[0083] The whole intervention analysis process forms a complete quality improvement closed loop. The system continuously tracks the follow-up performance of the trained nursing staff, compares the deviation rate before and after training, and adjusts the weight and content of different training modules dynamically. The nursing management end provides a multi-dimensional quality monitoring dashboard to show the overall operation compliance rate, training completion, skill improvement trend and other summary data of the department. These data support nursing managers to develop targeted team capability improvement plans and form a continuous improvement professional development mechanism.

[0084] The reliability of data collection and processing is guaranteed by multiple check mechanisms. The system performs integrity checks on all input nursing operation records to verify the continuity and logical reasonableness of the timestamps. Physiological monitoring data undergo signal quality assessment to automatically identify and exclude abnormal values caused by motion artifacts or device interference. The matching of key time nodes uses a bidirectional verification mechanism, both by searching for corresponding physiological data changes in nursing records and by tracing possible interventions from significant physiological fluctuations. This bidirectional tracing method improves the accuracy of event correlation and reduces false positives and false negatives.

[0085] The system interface design supports multi-angle data analysis needs. The clinical view focuses on the intervention response chain of a single patient, displaying the correlation between nursing operations, physiological changes, environmental factors, and other events in a timeline format. The management view provides department-level quality index summaries, supporting multi-dimensional filtering and comparative analysis by ward, shift, and personnel category. The training view integrates individual and team skill development trajectories, visualizing weaknesses and improvement progress. All views support drill-down queries, allowing users to progressively delve from summary data to specific operation details and original records.

[0086] The integration of the intervention analysis module with existing hospital information systems uses standardized interfaces. The system obtains patient basic information and historical data from the electronic medical record system through the HL7 protocol and retrieves relevant examination results from the medical imaging system through the DICOM standard. Nursing operation data are converted and mapped through specialized middleware to ensure accurate correlation of information from different sources. Analysis results are returned to the clinical system in the form of structured documents, embedded in the patient's electronic nursing record, forming a complete diagnosis and treatment evidence chain.

[0087] Through this fine implementation, the intervention analysis module realizes the complete closed-loop management from nursing operation to physiological response to quality improvement. The system not only identifies the surface operation deviation, but also deeply analyzes the root cause and influencing factors of the deviation. The evidence-based training program design makes the capacity improvement measures more targeted, and the application of virtual simulation technology creates a safe repeated practice environment. The multi-dimensional evaluation system objectively reflects the skill level of nursing staff, and the data-driven continuous improvement mechanism promotes the improvement of overall nursing quality. The operation of the whole module does not depend on subjective judgment, but through quantitative analysis and probabilistic reasoning, it provides objective and reliable decision support for clinical quality management.

[0088] In Example 4, the early warning execution module builds a personalized clinical skill reinforcement system through deep analysis of nursing operation deviation data. Taking the infusion rate adjustment deviation of vasoactive drugs in a cardiogenic shock rescue as an example, the system first analyzes the operation type code "VA-203" corresponding to the vasoactive drug dose adjustment operation. The electronic archive of the operator shows that the nursing staff completed the basic life support training in the past 12 months, but the advanced cardiovascular life support (ACLS) certification has expired for 8 months. The system further retrieves its operation record database within three years and finds that similar drug dose adjustment deviations have occurred 3 times in the past 6 months, all involving norepinephrine infusion scenarios.

[0089] The system establishes a skill defect analysis matrix to map the "VA-203" type deviation to 7 skill assessment dimensions. As shown in Table 1, part of the skill assessment data of the nursing staff is shown.

[0090] Table 1: Part of the skill assessment data of the nursing staff

[0091]

[0092] Based on the analysis results, the training construction subsystem dynamically configures the virtual reality training scene. For the drug calculation dimension, the system generates an interactive question bank containing multiple vasoactive drug concentration conversions, simulating the common drug specification changes during rescue. When the virtual patient's blood pressure drops sharply, the student needs to complete the concentration conversion from 50mg / 50ml to 2mg / ml within a limited time and correctly set the infusion pump parameters. The equipment operation dimension focuses on the coordinated use of multi-channel infusion pumps, simulating the complex scenario of simultaneously adjusting the infusion rates of norepinephrine and dopamine, and the system monitors the dose ratio of the two channels in real time to meet the requirements of the treatment guidelines.

[0093] Virtual scenario design incorporates clinical decision tree logic. Take a case of cardiogenic shock for example, the initial scenario presents a virtual patient with blood pressure 82 / 45 mmHg and heart rate 118 beats per minute. The learner needs to choose the category and initial dose of vasoactive drugs according to the real-time changing vital signs. The built-in physiological model produces dynamic responses based on each operation decision of the learner: correct drug selection and dose adjustment gradually improves the patient's condition, while the wrong decision leads to the deterioration of the patient's condition. Multiple decision branch points are set in the scenario, for example, when the learner chooses to use dopamine instead of norepinephrine, the system will trigger the pathological and physiological changes of increased myocardial oxygen consumption, guiding the learner to understand the theoretical basis of drug selection.

[0094] The program generation subsystem integrates multi-dimensional evaluation data to generate a customized training program for the nurse in the example. The program includes 8 hours of virtual reality hands-on training, focusing on emergency response and process execution. The ACLS protocol training module reproduces 10 typical arrhythmia scenarios, requiring the learner to correctly identify the rhythm on the virtual defibrillator and choose the synchronous / asynchronous mode. The system automatically adjusts the difficulty of the examination according to the historical passing rate, and for the nurse's process execution dimension with a 75% passing rate, it is set that the learner must complete 3 perfect process executions in a row to pass. The theoretical training part focuses on the pharmacological properties of vasoactive drugs, including 20 drug selection situational judgment questions based on real cases.

[0095] The training effect evaluation adopts a multi-time point evaluation mechanism. After completing the virtual reality training, the system arranges the nurse to perform physical operation assessment in the mannequin laboratory. The assessment scenario simulates the ICU environment and sets up a sudden blood pressure drop emergency, requiring the operator to independently complete the whole process from drug preparation to infusion rate adjustment. The evaluation indicators include operation time, dose calculation error rate, equipment setting accuracy, etc. Quantitative parameters, as well as the clinical thinking process at key decision points are recorded. The system compares the examination results with the baseline data before training, and generates an evaluation report containing the progress of each dimension.

[0096] The continuous tracking mechanism ensures the transformation of training effect. The system continuously monitors the actual clinical operation data of the nurse within 3 months after training. Each time the nurse performs vasoactive drug adjustment, the mobile terminal automatically pops up key operation point prompts and records the deviation of actual execution parameters from standard values. The system establishes a personal skill development curve and regularly analyzes the trend of operation accuracy. When it detects signs of degradation in a specific skill dimension, it automatically triggers the micro-training module to push targeted review content and short situational tests.

[0097] The management end supports team capability development monitoring. Nursing managers can view the overall skill matrix heat map of the department to intuitively find the weak links of the team. The system automatically generates a quarterly training analysis report, including the average compliance rate of each skill dimension, training needs distribution, resource input-output ratio, and other management indicators. For the ACLS protocol execution problem found in the example, the system suggests organizing a full-scene simulation drill for the team and marking the list of key personnel whose certification is about to expire.

[0098] Data interaction optimizes clinical workflow. The training system synchronizes with the hospital human resource management system in real time, automatically updating the certification status of nursing staff. When detecting that the ACLS certification is about to expire, it pushes a renewal reminder 3 months in advance and reserves training seats. The mobile application supports offline learning mode, allowing nursing staff to complete theoretical module learning during their spare time. Training progress data is integrated with the scheduling system to avoid scheduling shifts during important operation periods for personnel undergoing critical skill training.

[0099] The entire warning and execution process forms a multi-level improvement mechanism from individual to system. Individual operation deviations trigger personal capability improvement programs, and common problems are identified to guide team-level intervention measures. Continuous data tracking verifies the improvement effect. Virtual reality technology creates a safe trial-and-error environment, allowing nursing staff to repeatedly practice complex operations in a risk-free scenario. Dynamic difficulty adjustment mechanisms ensure that training is both challenging and achievable. The quantitative evaluation system eliminates subjective judgment bias and provides an objective reference for career development. The system is deeply integrated with existing hospital information platforms, making training management a natural extension of daily clinical work rather than an additional burden.

[0100] Example 5: Environmental calibration function of the vital sign monitoring module Through multi-sensor data fusion technology, environmental interference compensation is achieved for physiological parameter measurement values. A network of temperature and humidity sensors is deployed in the ward to collect environmental data at a frequency of minutes. The system establishes a dynamic correction model between environmental parameters and physiological indicators. When the temperature sensor detects that the ward temperature deviates from the standard value of 22 degrees Celsius, the system adjusts the raw blood pressure values obtained by the non-invasive blood pressure monitor according to the pre-set temperature-vessel constriction effect relationship. This correction considers the progressive effect of temperature changes on vessel tone and uses a nonlinear compensation algorithm to handle measurement deviations in sudden temperature environments. At the same time, the system monitors changes in relative humidity in the ward and activates a special compensation mode when humidity exceeds 60%. It considers the impact of changes in skin conductivity on ECG signal quality in high humidity environments.

[0101] Drug interference analysis function deeply integrates hospital pharmacy management system data flow. The system obtains the information of vasoactive drug infusion in the electronic medical order system in real time, including the current infusion rate and cumulative dose of drugs such as norepinephrine and dopamine. The pharmacokinetic model established for each type of drug calculates the expected effect of drug concentration on autonomic nervous system function. For example, for alpha receptor agonists, the system predicts the heart rate variability inhibition effect that may occur within 30-90 seconds after drug administration, and marks the time period that may be affected by the drug in the original heart rate variability data. The system maintains a drug effect time window database to record the typical time characteristics of drug onset, peak and decline under different administration methods, providing a pharmacological background reference for physiological parameter interpretation.

[0102] The data correction engine uses a multi-level verification mechanism to process the original physiological signals. The system first verifies the credibility of environmental sensor data, eliminating abnormal readings caused by sensor failure or temporary obstruction. Then the time alignment operation is performed to ensure that the environmental parameter changes and physiological index fluctuations have accurate time correspondence. In the drug effect compensation stage, the system distinguishes between intravenous bolus and continuous infusion administration modes and uses different compensation strategies. For fast bolus drugs, the system sets a dynamic compensation window before and after the administration time point, and the window width is automatically adjusted according to the drug characteristics. Continuous infusion drugs trigger continuous compensation mode, and the system updates compensation parameters in real time according to the infusion rate changes. The final output of the corrected data retains the dual record of the original measurement value and the corrected value for clinical personnel to compare and reference.

[0103] The drug safety monitoring module constructs a drug-physiological indicator dynamic response model. The system continuously analyzes the potassium concentration trend of patients using diuretics and establishes a prediction curve based on the dose and time function of drug administration. When the furosemide infusion rate and the blood potassium decline speed are detected to be outside the normal proportion range, the system starts the risk warning process. The warning mechanism uses a hierarchical triggering strategy to judge the risk level according to the combination of blood potassium absolute value and decline speed. The primary warning is triggered when the blood potassium is between 3.0-3.5 mmol / L and rapidly decreases, and the system displays warning information on the nursing terminal; the intermediate warning is activated when the blood potassium is lower than 3.0 mmol / L, and the system automatically locks the infusion pump and prompts to immediately supplement potassium; the high-level warning is for critical situations where the blood potassium is lower than 2.5 mmol / L, and the system notifies the medical team and generates an emergency consultation request at the same time.

[0104] The alternative therapy recommendation engine integrates a clinical guideline knowledge base. When the system detects a high-risk drug compatibility situation, it automatically retrieves alternative drug options from the in-hospital medication knowledge base. For furosemide-induced hypokalemia, the system recommends different potassium supplementation strategies such as 10% potassium chloride or potassium magnesium aspartate based on the patient's renal function and current electrolyte levels. The recommended plan includes detailed preparation methods, infusion rate ranges, and monitoring requirements to support nursing staff in quick execution. The system also provides associated monitoring recommendations, such as increasing the frequency of ECG ST segment monitoring or scheduling an emergency electrolyte review.

[0105] The environmental-drug-physiological closed-loop control system enables real-time intervention adjustments. When the environmental temperature continues to rise, causing the blood pressure correction value to show an upward trend, the system automatically adjusts the ward air conditioning parameters to maintain a stable measurement environment. For cases where significant physiological parameter changes are detected due to drug effects, the system dynamically adjusts the monitoring plan, such as increasing ECG leads or shortening non-invasive blood pressure measurement intervals. This adaptive monitoring strategy is personalized according to individual patient responses, avoiding a one-size-fits-all fixed monitoring mode. The system records logs of all automatic adjustment operations, including adjustment time, adjustment parameters, and adjustment basis, forming a complete closed-loop control record.

[0106] The clinical decision support interface presents a multi-dimensional data correlation view. The system designs a special environmental-drug influence dashboard to display temperature change curves, drug infusion rate curves, and physiological parameter curves before and after correction in a time axis format. Clinical staff can select different display combinations through interactive controls to visually observe the combined effects of environmental factors and drug interventions on physiological indicators. Key safety warning information uses a color coding system, with different background colors from light yellow to deep red highlighted according to the urgency level. The interface retains the manual override authority of professional clinical staff, allowing manual adjustment of system parameters or temporary disabling of specific compensation functions in special cases.

[0107] The system calibration mechanism ensures long-term operation accuracy. Regularly perform sensor cross-validation procedures to compare the consistency of temperature and humidity sensor readings from different locations, identify possible device drift. The drug influence model is updated monthly, incorporating the latest pharmacological research data and clinical medication feedback. The system maintenance team regularly reviews automatic correction cases and samples to check the clinical reasonableness of the correction algorithm, and manually adjusts the compensation coefficients if necessary. This quality control mechanism combining automatic calibration and manual review ensures the stability and reliability of the system in various clinical environments.

[0108] Deep integration with hospital infrastructure expands system function boundaries. The system interfaces with building management systems to obtain air conditioning operation status and air flow organization data, and to predict temperature regulation response delays. Direct connection with laboratory information systems obtains key test results such as electrolyte and blood gas analysis in real time, and verifies the biochemical basis of physiological parameter changes. Integration with nursing call systems activates bedside call lights when advanced alarms are triggered, attracting the attention of nursing staff. This cross-system collaboration mode breaks down information silos and builds a comprehensive patient safety protection network.

[0109] Through this fine implementation, the system realizes intelligent compensation for environmental interference and drug effects, and improves the clinical reliability of raw physiological data. Multi-parameter fusion analysis reveals risk patterns that cannot be found by traditional single monitoring, and early warning mechanisms save valuable time for clinical intervention. Closed-loop control characteristics enable the monitoring system to adapt to complex and changing clinical environments and maintain stable monitoring quality. An intuitive decision support interface helps clinical staff quickly understand the physiological and pathological significance behind the data and make more accurate clinical judgments. The entire system operates without interfering with the regular diagnosis and treatment process, silently enhancing every link of the patient safety protection system.

[0110] It should be noted that the relational terms herein such as first and second and the like are used solely to distinguish one entity or action from another, without necessarily requiring or implying any such actual relationship or order between or among such entities or actions. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0111] Although embodiments of the present application have been shown and described, it is to be understood that various modifications, substitutions, replacements and variations can be made to these embodiments without departing from the principles and spirit of the present application, and the scope of the present application is defined by the appended claims and their equivalents.

Claims

1. A cloud-based cardiology nursing information monitoring system, characterized in that, The system comprises: The vital sign monitoring module collects the multi-dimensional physiological parameters of the patient in the cardiology department, integrates the real-time data streams of the wearable device and the bedside monitor, continuously tracks the heart rate variability and blood pressure fluctuation curve of the patient, calculates the abnormal offset amount of the respiratory rate, and generates a dynamic vital sign set in combination with the oxygen saturation change trend; The health assessment module receives the dynamic vital sign set, identifies the abnormal fluctuation nodes in the vital sign data, matches the health event records in the patient's historical medical records, quantifies the correlation strength between the current vital signs and the typical pathological patterns, evaluates the deviation degree of the patient's health status, and generates a multi-dimensional health risk indicator; The resource allocation module analyzes the real-time occupancy state of the nursing resources based on the multi-dimensional health risk indicator, calculates the emergency weight of different nursing tasks, dynamically allocates the work path of medical staff and the use sequence of medical equipment, and generates a hierarchical nursing scheduling scheme; The intervention analysis module executes the hierarchical nursing scheduling scheme, compares the changes in vital sign data before and after resource allocation, detects the response delay time of abnormal vital signs, locates the execution deviation nodes of nursing intervention measures, and generates a nursing effect traceability report; The resource allocation module comprises: The task analysis submodule decomposes the heart function classification label in the multi-dimensional health risk indicator, maps the corresponding nursing operation list of different grades, identifies the inventory status of emergency medicines and the occupancy status of respirators, and calculates the device dependency coefficient of each nursing operation; The path optimization submodule receives the nursing operation list and device dependency coefficient, constructs a real-time topological graph of medical staff location information, simulates the task completion time delay rate under different paths, and iteratively optimizes the execution order of the nursing operation sequence; The scheduling generation submodule integrates the optimized nursing operation sequence and device allocation scheme, coordinates the use time conflict of intravenous infusion pumps and defibrillators, and generates the hierarchical nursing scheduling scheme containing operation time window and device handover node.

2. The cloud computing based cardiology nursing information monitoring system according to claim 1, wherein, The vital sign monitoring module comprises: The multi-source fusion submodule synchronously receives heterogeneous data streams of electrocardiogram monitor, pulse oximeter and non-invasive blood pressure meter, calibrates the sampling time stamps of different devices, extracts the extreme points of physiological parameters within five consecutive minutes, calculates the synchronicity index of parameter fluctuation frequency, and generates a time-aligned vital sign data set; The anomaly detection submodule analyzes the time-aligned vital sign data set, divides the reference parameter interval of the diurnal cycle, monitors the duration of the real-time parameter exceeding the interval threshold, counts the occurrence density of abnormal events per unit time, and generates a vital sign abnormality probability matrix; The risk integration submodule calls the vital sign abnormality probability matrix, associates the drug metabolism time window in the patient's medication record, calculates the correlation coefficient of abnormal vital signs and drug concentration, integrates external interference factors of environmental temperature and humidity sensors, and generates the dynamic vital sign set. 3.The cloud computing based cardiology nursing information monitoring system according to claim 1, wherein, The health assessment module comprises: The pathological matching submodule extracts the heart rate shock parameter in the dynamic vital sign set, compares the waveform feature library of typical heart failure cases, calculates the spectral difference degree between the current waveform and the pathological reference waveform, and identifies the occurrence position and duration interval of premature beat events; The risk quantification submodule receives the spectrum difference and early beat event data, loads the myocardial enzyme historical value in the patient electronic medical record, constructs a dynamic change curve of the current myocardial injury index, and quantifies the proximity of the curve slope and the critical value; The index generation submodule integrates the dynamic change curve of the myocardial injury index and the pathological matching result, weighted calculates the priority order of the risk of organ failure, and generates the multi-dimensional health risk index including the heart function classification label and the organ risk weight.

4. The cloud computing based cardiology nursing information monitoring system according to claim 1, wherein, The intervention analysis module comprises: The response monitoring submodule captures the real-time sign data stream after the implementation of the grading nursing scheduling scheme, marks the corresponding relationship between the diuretic administration time point and the blood pressure drop event, and counts the delay length of the sign parameter returning to the safety threshold; The deviation positioning submodule compares the nursing operation preset standard with the time stamp of the execution record, identifies the operation node of the intravenous bolus speed deviating from the standard value, and associates the deviating operation with the blood potassium concentration mutation event in the adjacent time window; The traceability generation submodule integrates the delay length and the occurrence position of the blood potassium concentration mutation event, constructs a causal chain model of the nursing measures and the sign response, and generates the nursing effect traceability report including the key intervention node execution deviation record.

5. The cloud computing based cardiology nursing information monitoring system according to claim 4, wherein, The system further comprises: The early warning execution module receives the key intervention node execution deviation record in the nursing effect traceability report, calls the operation history database of the corresponding medical staff, matches the validity period state of the emergency skill certification certificate, and generates an individualized skill reinforcement training scheme; The individualized skill reinforcement training scheme includes the operation procedure simulating the emergency scene, the equipment operation standard video tutorial, and the examination and evaluation index.

6. The cloud computing based cardiology nursing information monitoring system according to claim 5, wherein, The early warning execution module comprises: The capability evaluation submodule analyzes the operation type of the key intervention node execution deviation record, retrieves the training completion rate data in the medical staff archive, and calculates the association strength of the operation error and the skill certificate update time; The training construction submodule configures a virtual reality simulation cardiogenic shock scene based on the operation type association strength result, and dynamically generates a correction training module of infusion rate adjustment error; The scheme generation submodule integrates the virtual reality simulation scene and the correction training module, superimposes the operation weak point analysis report in the historical examination, and generates the individualized skill reinforcement training scheme including the training class hour allocation and the examination threshold. 7.The cloud computing based cardiology nursing information monitoring system according to claim 1, wherein, The sign monitoring module further comprises: The environment calibration submodule obtains the real-time reading of the ward temperature and humidity sensor, calculates the influence coefficient of temperature fluctuation on vasoconstrictor, and corrects the environmental interference deviation amount of blood pressure monitoring value; The drug interference analysis submodule associates the blood vessel active drug infusion rate in the electronic medical order system, quantifies the lag response time of drug concentration peak and heart rate variation, and generates a drug influence compensation parameter; The data correction submodule fuses the environmental interference deviation amount and the drug influence compensation parameter, reconstructs the reference value of the original sign data, and updates the blood pressure fluctuation curve in the dynamic vital sign set. 8.The cloud computing based cardiology nursing information monitoring system according to claim 7, wherein, The path optimization submodule performs operations including: A real-time position tracking unit obtains indoor positioning beacon data worn by medical staff, and constructs a ward three-dimensional space coordinate mapping model; The task conflict detection unit identifies the geographical position radius of multiple emergency tasks in the same period, calculates the resource competition probability of the path intersection area; The dynamic adjustment unit reassigns the execution subject of the nursing task based on the resource competition probability result to avoid overlapping conflicts of the equipment transportation path. 9.The cloud computing based cardiology nursing information monitoring system according to claim 1, wherein, The system further comprises: The medication safety monitoring module receives the blood potassium concentration data in the dynamic vital sign set, associates the diuretic infusion rate parameter being executed by the electronic medical order system, identifies the potential risk time window of the drug incompatibility combination, and generates a high-risk drug compatibility warning instruction; The high-risk drug compatibility warning instruction triggers an automatic locking mechanism of the infusion pump and synchronously pushes a replacement treatment scheme to the mobile nursing terminal.

Images