A dynamic hemodynamic monitoring and support system for critically ill ICU patients

Abstract

This invention discloses a dynamic hemodynamic monitoring and support system for critically ill ICU patients, relating to the field of medical monitoring technology. Through a multi-parameter synchronous acquisition module, it achieves simultaneous and linked acquisition of hemodynamic parameters and coagulation function indicators. A data preprocessing module performs precise filtering, outlier removal, and missing value completion on the acquired raw data. A dynamic coupling analysis module constructs a quantitative blood flow-coagulation correlation model, which can dynamically adjust parameter warning thresholds based on the patient's individual characteristics and condition, accurately calculate coupling risk values, and determine risk types and levels. An intelligent intervention and control module formulates differentiated treatment plans based on coupling risks and achieves closed-loop intervention through a feedback correction mechanism. A display, alarm, and data storage module realizes the visualization of multi-dimensional information and full-process data retention, achieving dynamic coupling monitoring and intelligent support for hemodynamics and coagulation function.

Description

Technical Field

[0001] This invention relates to the field of medical monitoring technology, specifically to a dynamic hemodynamic monitoring and support system for critically ill ICU patients. Background Technology

[0002] In the field of intensive care medicine, ICU patients often face the dual challenges of circulatory and coagulation disorders due to the severity and complexity of their conditions. Stable circulatory function is crucial for maintaining blood perfusion and normal metabolism of the whole body organs, while normal coagulation function is related to whether patients will develop life-threatening complications such as bleeding or thrombosis. Therefore, real-time and accurate monitoring of hemodynamic parameters and coagulation function indicators of critically ill patients, and timely and effective intervention support based on the monitoring results, are key links to improve the success rate of treatment and prognosis of critically ill patients.

[0003] Existing monitoring methods have several significant shortcomings. First, hemodynamic monitoring and coagulation function monitoring systems operate independently, lacking effective data linkage and analysis capabilities. This makes it difficult to establish a dynamic correlation between the two, hindering a holistic understanding of patient condition changes and resulting in incomplete and inaccurate assessments. Second, data processing only provides basic noise reduction, lacking precision in identifying outliers and employing simplistic methods for completing missing values, which can interfere with subsequent analysis and affect accurate diagnosis. Third, intervention strategies rely solely on single-dimensional parameters without considering coagulation-blood flow coupling risks, increasing the risk of serious complications such as thrombosis or bleeding and raising treatment risks. Fourth, alarm mechanisms only target single parameter thresholds, lacking tiered warning capabilities for coupling risks. These shortcomings significantly hinder the timeliness and accuracy of treatment decisions, severely impacting the treatment outcomes for critically ill patients. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a dynamic monitoring and support system for hemodynamics in ICU critically ill patients. This system achieves simultaneous and linked acquisition of hemodynamic parameters and coagulation function indicators through a multi-parameter synchronous acquisition module, solving the problem of fragmented monitoring dimensions in existing systems. The data preprocessing module performs precise filtering, outlier removal, and missing value completion on the acquired raw data, effectively improving data accuracy. The dynamic coupling analysis module constructs a quantitative blood flow-coagulation correlation model, which can dynamically adjust parameter warning thresholds based on the patient's individual characteristics and condition, accurately calculate coupling risk values, and determine risk types and levels, providing a scientific basis for personalized intervention. The intelligent intervention and control module formulates differentiated treatment plans based on coupling risks and achieves closed-loop intervention through a feedback correction mechanism, significantly improving the safety and effectiveness of treatment. The display, alarm, and data storage module realizes the visualization of multi-dimensional information and full-process data retention, facilitating real-time monitoring of patient conditions and system operation by medical staff, achieving dynamic coupling monitoring and intelligent support for hemodynamics and coagulation function.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a dynamic hemodynamic monitoring and support system for critically ill ICU patients, the system comprising: Multi-parameter synchronous acquisition module: integrates hemodynamic acquisition unit and coagulation function acquisition unit, synchronously and frequently acquires patient hemodynamic parameters and periodically acquires coagulation function indicators with bedside detection equipment through adapter sensors, and transmits the raw data to data preprocessing module; Data preprocessing module: Cleans and standardizes the raw data through filtering, outlier identification and removal, and interpolation completion techniques, outputting continuous and accurate standardized data and transmitting it to the dynamic coupling analysis module; Dynamic Coupling Analysis Module: Based on an individual characteristic database, the module calculates coupling risk values ​​and determines risk types and levels by dynamically adjusting early warning thresholds and constructing a blood flow-coagulation correlation model. The analysis results are then transmitted to the intelligent intervention and control module. Intelligent intervention and control module: Based on the intervention strategy library, it matches differentiated treatment plans, generates and sends control commands for treatment equipment, and corrects the commands through a timed feedback and correction mechanism to ensure that the intervention effect is consistent with the changes in the patient's condition; Display alarm and data storage module: Real-time display of monitoring data, risk status and intervention instructions through interactive terminal, hierarchical audible and visual alarms based on coupled risk values, automatic storage of full-process data, and support for reset operation and data retention.

[0006] Furthermore, the hemodynamics acquisition unit consists of an arterial pressure sensor, a central venous pressure sensor, a cardiac output monitoring sensor, a peripheral blood oxygen saturation sensor, and a body temperature sensor, which collects systolic blood pressure, diastolic blood pressure, mean arterial pressure, central venous pressure, cardiac output, stroke volume, peripheral vascular resistance, blood oxygen saturation, and core body temperature parameters in real time. The coagulation function acquisition unit is equipped with a thromboelastography instrument, a platelet function analyzer, a thrombin generation rate detector, and a fibrinogen concentration detector. It is connected to the patient's peripheral vein through a disposable sterile blood collection tubing to collect thrombin generation rate, fibrinogen level, platelet aggregation rate, and thromboelastography parameters, achieving synchronous acquisition of blood flow and coagulation indicators.

[0007] Furthermore, the data preprocessing module employs a filtering algorithm to reduce the impact of signal noise on data accuracy, taking into account the susceptibility of raw hemodynamic parameters to interference. For coagulation function data, an outlier identification algorithm is used to filter and remove erroneous data that deviates from the normal range. For data gaps that occur during data transmission or acquisition, interpolation is used to ensure the integrity and continuity of the data sequence.

[0008] Furthermore, the dynamic coupling analysis module incorporates an individual characteristic database and a coupling analysis model. The individual characteristic database pre-stores basic patient information, as well as baseline hemodynamic parameters and coagulation function indicators upon admission, providing a basis for personalized analysis. It dynamically adjusts the warning thresholds of each parameter based on the deviation between the patient's real-time parameters and baseline data, and the time of disease progression. Simultaneously, it constructs an association model and trains it based on clinical data, calibrating the weight coefficients and correction constants of each core association indicator. Hemodynamic parameters and coagulation function indicators are selected as core association indicators, and differentiated weights are assigned to each indicator according to the patient's disease type. The coupling risk value is then calculated in conjunction with the correction constant. The positive or negative value of the coupling risk value determines the bleeding or thrombosis risk type, and the absolute value is used to classify the risk level. After the analysis is completed, the risk type, level, and related parameter data are transmitted to the intelligent intervention and control module.

[0009] Furthermore, the dynamic coupling analysis module dynamically adjusts the warning thresholds of each parameter based on the deviation between the patient's real-time parameters and baseline data, as well as the time of disease progression. The adjustment formula is as follows: ,in, It is the first The lower limit of the parameter warning threshold at any given time, i.e., the minimum safe threshold after dynamic adjustment. These are the baseline values ​​of the corresponding parameters when the patient is admitted, stored in the individual characteristic database of the dynamic coupling analysis module, and serve as the basis for threshold adjustment. This is the disease condition impact coefficient, with a value ranging from 0.1 to 0.3, which is automatically matched by the dynamic coupling analysis module based on the patient's disease type. It is the time decay coefficient, with a fixed value of 0.05 / h, which represents the decay rate of the disease influence coefficient as the treatment time increases (i.e., the longer the treatment time, the weaker the interference of the disease on the threshold, and the closer the threshold is to the baseline value). It indicates the progression of the disease, specifically the cumulative duration of monitoring and treatment received by the system since the patient's admission, which is updated in real time by the dynamic coupling analysis module.

[0010] Furthermore, the dynamic coupling analysis module constructs an association model, the model formula of which is: ,in, This is a coupled risk value used to determine the type and level of risk. Negative values ​​indicate bleeding risk, while positive values ​​indicate thrombosis risk. The larger the absolute value, the higher the risk level. , These are the core hemodynamic parameters, among which Corresponding mean arterial pressure, Corresponding cardiac output; , , The core coagulation function indicators, among which Corresponding thrombin generation rate, Corresponding fibrinogen level, Corresponding platelet aggregation rate; , It is the differential weighting of hemodynamic parameters. , , It is the differentiated weighting of coagulation function indicators. It is a correction constant, which is a preset system calibration value.

[0011] Furthermore, the dynamic coupling analysis module is trained based on clinical data, calibrating the weight coefficients and correction constants of each core correlation indicator. The specific training steps are as follows: First, clinical data of ICU critically ill patients are collected, including corresponding hemodynamic parameters, coagulation function indicators, and clinical outcomes. After cleaning and removing outliers and standardizing the data, it is divided into training and validation sets according to the proportion. Using hemodynamic parameters and coagulation function indicators as input features and bleeding / thrombosis labels corresponding to clinical outcomes as output labels, a multivariate linear regression model framework is constructed. The weight coefficients and correction constants of each feature are iteratively optimized using the least squares method through the training set data to achieve the best match between the coupling risk value output by the model and the actual clinical risk status. Then, the model performance is evaluated using the validation set data, with accuracy, sensitivity, and specificity as core indicators. If the performance does not reach the preset threshold, the feature dimensions are readjusted or the algorithm parameters are optimized and the model is trained again. After training, the model is fixed to the dynamic coupling analysis module.

[0012] Furthermore, the aforementioned It is the coupling risk value, when A value <0 indicates a bleeding risk; a negative value indicates that coagulation function-related indicators are dominant, or hemodynamic parameters (indicating bleeding tendency; the smaller the value, the higher the bleeding risk); when A value >0 indicates a risk of thrombosis. A positive value indicates that coagulation function-related indicators are dominant, or that hemodynamic parameters suggest a tendency to thrombosis. The higher the value, the higher the risk of thrombosis. No significant risk: when... =0 or close to 0, that is, when the absolute value is <5, it is determined that there is no significant risk of bleeding / thrombosis, and hemodynamics and coagulation function are in a relatively balanced state.

[0013] Furthermore, the intelligent intervention and control module has a built-in intervention strategy library and correction mechanism. The intervention strategy library pre-stores differentiated intervention rules for different disease types, which can match the corresponding treatment plan according to different disease states and coupling risk results. Based on the risk judgment results transmitted by the dynamic coupling analysis module, the corresponding treatment plan is accurately matched from the strategy library to generate control instructions including fluid infusion rate and drug administration rate. These instructions are transmitted to the infusion pump and drug administration device in the ICU through the adapter bus. The module also has a timed feedback correction capability, which collects the adjusted parameters and recalculates the coupling risk value. At the same time, the execution status of the instructions is synchronously transmitted to the display alarm and data storage module.

[0014] Furthermore, the intelligent intervention and control module collects the adjusted parameters and recalculates the coupling risk value. If the risk value returns to the safe range, the current treatment plan is maintained. If the risk value does not return to the safe range, the system will intelligently correct the instruction according to the magnitude of the risk change. If the risk value has decreased significantly compared to before the adjustment but has not reached the safe range, the current adjustment direction is maintained and the adjustment magnitude is optimized. If the risk value does not improve significantly or even increases, the system will immediately switch to the corresponding alternative treatment plan in the strategy library.

[0015] Compared with existing technologies, this dynamic hemodynamic monitoring and support system for critically ill ICU patients has the following advantages: This invention constructs a dynamic coupling mechanism between hemodynamics and coagulation function, enabling deep integration and collaborative analysis of multi-dimensional data. This effectively solves the problems of functional fragmentation and poor data linkage in existing monitoring systems. By establishing a quantitative coupling model, the weights of hemodynamic and coagulation function indicators can be dynamically allocated according to the patient's condition, accurately determining the type and level of bleeding or thrombosis risk, and providing a scientific basis for personalized intervention. Simultaneously, through the closed-loop feedback mechanism of the intelligent intervention and control module, combined with the coupled risk value, the treatment plan is adjusted in real time, significantly improving the safety and effectiveness of treatment. Finally, through multi-parameter synchronous acquisition and full-process data retention, the synchronization, comprehensiveness, and traceability of monitoring data are ensured, significantly improving the accuracy of critical care patient monitoring and the scientific nature of clinical decision-making.

[0016] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0018] Figure 1 A structural block diagram of a dynamic hemodynamic monitoring and support system for critically ill ICU patients; Figure 2 A flowchart of a dynamic hemodynamic monitoring and support system for critically ill ICU patients; Figure 3 This is a flowchart of a dynamic coupling analysis module for a hemodynamic monitoring and support system for critically ill ICU patients. Detailed Implementation

[0019] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0020] This invention provides a dynamic hemodynamic monitoring and support system for critically ill ICU patients, such as... Figure 1As shown, the system includes a multi-parameter synchronous acquisition module, a data preprocessing module, a dynamic coupling analysis module, an intelligent intervention and control module, and a display, alarm, and data storage module. The multi-parameter synchronous acquisition module enables the synchronous and linked acquisition of hemodynamic parameters and coagulation function indicators, solving the problem of fragmented monitoring dimensions in existing systems. The data preprocessing module performs precise filtering, outlier removal, and missing value completion on the acquired raw data, effectively improving data accuracy. The dynamic coupling analysis module constructs a quantitative blood flow-coagulation correlation model, which can dynamically adjust parameter warning thresholds based on the patient's individual characteristics and condition, accurately calculate coupling risk values, and determine risk types and levels, providing a scientific basis for personalized intervention. The intelligent intervention and control module formulates differentiated treatment plans based on coupling risks and combines a feedback correction mechanism to achieve closed-loop intervention, significantly improving the safety and effectiveness of treatment. The display, alarm, and data storage module realizes the visualization of multi-dimensional information and full-process data retention, facilitating real-time monitoring of patient conditions and system operation by medical staff, achieving dynamic coupling monitoring and intelligent support of hemodynamics and coagulation function. The multi-parameter synchronous acquisition module integrates a hemodynamic acquisition unit and a coagulation function acquisition unit. Through the adapter sensor and bedside detection equipment, it synchronously and frequently acquires the patient's hemodynamic parameters and periodically acquires coagulation function indicators, and transmits the raw data to the data preprocessing module.

[0021] Hemodynamics Acquisition Unit: Adopting a modular design, this unit integrates multiple medical-grade precision sensors, including an arterial pressure sensor, a central venous pressure sensor, a cardiac output monitoring sensor, a peripheral blood oxygen saturation sensor, and a body temperature sensor. The arterial pressure sensor is connected to the corresponding artery via an arterial puncture catheter; the central venous pressure sensor is connected via a central venous catheter; the cardiac output monitoring sensor utilizes pulse indication continuous cardiac output monitoring technology; the peripheral blood oxygen saturation sensor is clipped to the patient's fingertip; and the core body temperature sensor uses an esophageal or bladder temperature probe. Each sensor collects its corresponding hemodynamic parameters, which are then converted from analog to digital and transmitted in real-time to the data preprocessing module as digital signals.

[0022] Coagulation function collection unit: Equipped with thromboelastography, platelet function analyzer, thrombin generation rate analyzer, fibrinogen concentration analyzer, etc. All devices establish access to the patient's peripheral vein through disposable sterile blood collection tubing. The tubing has an anticoagulant coating to prevent blood clotting during collection from affecting the test results. The collection process is carried out according to the set sequence. The collected blood samples are injected into the reaction pool of each testing device. Each device detects the corresponding coagulation function-related indicators. After the test is completed, the data and hemodynamic parameters are synchronously transmitted to the data preprocessing module.

[0023] Data preprocessing module: Cleans and standardizes the raw data through filtering, outlier identification and removal, and interpolation completion techniques, outputting continuous and accurate standardized data and transmitting it to the dynamic coupling analysis module.

[0024] Hemodynamic data filtering: Hemodynamic parameters are easily affected by factors such as patient limb movement and electromagnetic interference from equipment during acquisition. An appropriate filtering algorithm is used to suppress noise. This algorithm can dynamically adjust the filtering gain, effectively filter high-frequency interference signals, and at the same time retain the true trend of parameter changes. Outlier removal of coagulation function data: An outlier identification method that combines clinically reasonable ranges is adopted. First, the coagulation function index data is initially screened to mark the data that deviates from the normal range. Then, it is verified again by combining the normal reference range recognized by clinical experience to remove the confirmed erroneous data. For the outliers that have been screened and confirmed, the system automatically records their collection time and device number. Data gap interpolation completion: To address data gaps caused by data transmission delays or brief equipment malfunctions, an appropriate interpolation method is selected based on the data type. For frequently acquired hemodynamic parameters, linear interpolation is used, utilizing adjacent valid data points before and after the gap to calculate the gap value, ensuring the smoothness of the data sequence. For periodically acquired coagulation function indicators, the corresponding interpolation method is selected based on the gap duration to ensure the integrity of the data sequence. The processed standardized data is sent to the dynamic coupling analysis module via an encrypted transmission protocol.

[0025] The dynamic coupling analysis module, based on an individual characteristic database and a coupling analysis model, dynamically adjusts warning thresholds, calculates coupling risk values, and determines risk types and levels. By deeply integrating individual patient characteristics and clinical data, it ensures the personalization and accuracy of the analysis results. Figure 3 As shown, the specific implementation process is as follows: Individual characteristic database construction and maintenance: The database pre-stores the patient's basic information, hemodynamic baseline parameters and coagulation function baseline indicators upon admission. The basic information includes key contents such as disease diagnosis, past medical history, and medication history, providing a comprehensive basis for personalized analysis. The baseline parameters and indicators are determined by the standardized data collected for the first time after admission, and are solidified and stored after review and confirmation, serving as the basic benchmark for subsequent analysis. The database supports a dual update mechanism. On the one hand, when a patient's condition undergoes significant changes, the system automatically triggers the baseline update process to re-collect and store new baseline data. On the other hand, doctors can manually initiate a baseline update request based on clinical judgment to ensure that the data always reflects the patient's latest condition. At the same time, the database uses encrypted storage to strictly protect the patient's medical privacy and security. Dynamic adjustment of warning threshold: The threshold is dynamically adjusted based on two dimensions: the deviation between the patient's real-time parameters and baseline data, and the time of disease progression. The core logic is to allow the warning threshold to adaptively optimize with the patient's treatment progress and changes in their condition. The adjustment formula is as follows: ,in, It is the first The lower limit of the parameter warning threshold at any given time, i.e., the minimum safe threshold after dynamic adjustment. These are the baseline values ​​of the corresponding parameters when the patient is admitted, stored in the individual characteristic database of the dynamic coupling analysis module, and serve as the basis for threshold adjustment. This is the disease condition impact coefficient, with a value ranging from 0.1 to 0.3, which is automatically matched by the dynamic coupling analysis module based on the patient's disease type. This is the time decay coefficient, fixed at 0.05 / h, which characterizes the rate at which the disease's influence coefficient decays as treatment time increases. In other words, the longer the treatment time, the weaker the interference of the disease on the threshold, and the closer the threshold is to the baseline value. The threshold reflects the progression of the patient's condition, specifically the cumulative duration of monitoring and treatment provided by the system since admission. This data is updated in real-time by the dynamic coupling analysis module. For patients with acute onset or significant fluctuations in their condition, the initial threshold adjustment range better reflects the urgency of the illness. As treatment progresses and the condition gradually stabilizes, the interference of the condition on the threshold weakens, and the threshold gradually approaches the baseline value. This avoids misjudgments or missed judgments caused by fixed thresholds. During the adjustment process, the system correlates the patient's treatment measures, such as medication use and surgical interventions, in real time. If the patient's condition shows a significant turning point after receiving critical treatment, the threshold adjustment rate will be optimized accordingly to ensure the timeliness and accuracy of the warnings.

[0026] Coupled analysis model construction and training: Model Construction: A coupling analysis model is constructed using a multiple linear regression model, and its formula is as follows: ,in, This is a coupled risk value used to determine the type and level of risk. Negative values ​​indicate bleeding risk, while positive values ​​indicate thrombosis risk. The larger the absolute value, the higher the risk level. , These are the core hemodynamic parameters, among which Corresponding mean arterial pressure, Corresponding cardiac output; , , The core coagulation function indicators, among which Corresponding thrombin generation rate, Corresponding fibrinogen level, Corresponding platelet aggregation rate; , It is the differential weighting of hemodynamic parameters. , , It is the differentiated weighting of coagulation function indicators. The calibration constant is a preset system calibration value. The core indicators were selected through clinical relevance analysis. Mean arterial pressure and cardiac output were determined as core hemodynamic parameters from hemodynamic parameters and coagulation function indicators, while thrombin generation rate, fibrinogen level, and platelet aggregation rate were determined as core coagulation function indicators. The model fully considers individual patient differences and assigns differentiated weights to each core indicator according to the type of disease, highlighting the impact of key indicators on risk assessment. At the same time, a preset calibration constant is introduced to balance the model calculation bias. Finally, the coupled risk value is obtained through multi-indicator fusion calculation. Model Training: Training data comes from clinical cases of critically ill patients in the ICU, covering case data of different age groups, disease types, and treatment outcomes to ensure the model's universality and adaptability. After data collection, it first undergoes strict quality review and cleaning to remove abnormal data caused by instrument malfunction or operational errors, and then undergoes standardization to unify the data scale. Subsequently, it is divided into training and validation sets according to a scientific ratio. The core parameters and indicators are used as input features, and the bleeding / thrombosis labels corresponding to the clinical outcomes are used as output labels. The model is constructed and the weight coefficients and correction constants are iteratively optimized through training set data. A quality control mechanism is established during the training process to regularly evaluate the matching degree between the model output results and the actual clinical situation. The model performance is evaluated from core dimensions such as accuracy, sensitivity, and specificity using validation set data. If the preset standards are not met, the feature dimensions are readjusted or the algorithm parameters are optimized, and the model is repeatedly trained until it meets the requirements of clinical application. Finally, the optimized model is solidified into this module. Risk assessment logic: The positive or negative value of the coupled risk value directly corresponds to the risk type. A negative value clearly indicates bleeding risk, while a positive value indicates thrombosis risk. The absolute value of the risk value intuitively reflects the risk level. The larger the absolute value, the higher the risk level, which helps medical staff quickly determine the urgency of the condition. When the coupled risk value is 0 or close to 0, it indicates that the patient's hemodynamics and coagulation function are in a relatively balanced state, with no significant bleeding or thrombosis risk. For different risk levels, the system sets differentiated analysis frequencies. In low-risk conditions, the analysis is performed at the regular frequency, while in medium- and high-risk conditions, the analysis frequency is increased to capture changes in the condition in real time. After the analysis is completed, the system not only outputs the risk type and level, but also simultaneously marks abnormal core indicators and deviations, providing clear guidance for subsequent intervention and treatment. Relevant data is transmitted to the intelligent intervention and control module in real time.

[0027] The intelligent intervention and control module, based on the risk assessment results of the dynamic coupling analysis module, achieves the matching and precise execution of differentiated treatment plans, and ensures the intervention effect through a timed feedback and correction mechanism. The specific implementation is as follows: Intervention strategy library construction: The strategy library pre-stores differentiated intervention rules for different disease types, risk types, and risk levels. Each rule clearly defines the core content of the corresponding treatment plan, including key parameters such as fluid replacement type, fluid replacement rate, drug type, dosage, and administration method. All intervention rules are formulated with reference to the latest clinical diagnosis and treatment guidelines and multi-center clinical practice experience. The strategy library supports flexible updates and personalized adjustments. Clinicians can modify and supplement the intervention rules according to the hospital's diagnosis and treatment characteristics and the individual patient's condition. All modification operations are automatically recorded for easy traceability and review. Treatment plan matching and instruction generation: After receiving the risk assessment results transmitted by the dynamic coupling analysis module, the intelligent intervention and control module quickly screens the optimal treatment plan. During the matching process, it comprehensively considers the patient's underlying diseases, medication history, allergy history and other individual information to avoid plan conflicts. After the treatment plan is determined, the system automatically generates control instructions containing key parameters such as fluid infusion rate and drug administration rate. The control instructions adopt a standardized communication protocol to ensure compatibility with treatment equipment such as infusion pumps and drug delivery devices in the ICU. They are accurately transmitted to the corresponding devices through the adapter bus to ensure the accuracy of instruction execution. A timed feedback correction mechanism is established: a fixed feedback correction cycle is set. After each instruction is executed, the system automatically collects the patient's adjusted hemodynamic parameters and coagulation function indicators, and transmits them to the data preprocessing module and dynamic coupling analysis module to recalculate the coupling risk value. If the risk value returns to the safe range, the current treatment plan is maintained; if the risk value does not return to the safe range, the system will intelligently correct the instruction based on the magnitude of the risk change. If the risk value has significantly decreased compared to before the adjustment but has not reached the safe range, the current adjustment direction is maintained and the adjustment magnitude is optimized; if the risk value does not improve significantly or even increases, the system immediately switches to the corresponding alternative treatment plan in the strategy library to avoid delaying the patient's condition. After the instruction is corrected, it is synchronized in real time to the treatment equipment, display alarm, and data storage modules to ensure consistency of information across all modules.

[0028] The alarm and data storage module displays monitoring data, risk status and intervention commands in real time through the bedside interactive terminal. It implements graded audible and visual alarms according to the coupled risk value, and automatically stores the entire process data. It also supports reset operations and data retention. Data visualization display: The bedside interactive terminal displays the patient's basic information, monitoring parameters (including real-time values ​​and change curves), coupled risk values, risk types and levels, current treatment plans and instruction execution status in real time. The interface adopts a partitioned design with clear functions for each area. Abnormal parameters are marked with eye-catching colors, and risk levels are presented intuitively through different colored icons, making it convenient for medical staff to quickly view key information and improve work efficiency. Tiered audible and visual alarms: Differentiated audible and visual alarms are triggered based on the coupled risk level. At low risk, the screen flashes a specific color to alert without causing sound interference. At medium risk, the screen flashes a specific color and is accompanied by intermittent buzzing to remind medical staff to pay attention. At high risk, the screen flashes a specific color continuously and is accompanied by a high-frequency continuous buzzing, simultaneously triggering the bedside call bell and the nurse station alarm system to ensure that medical staff can respond immediately. After the alarm is triggered, the system automatically records the alarm time, risk level, and relevant abnormal indicators. The alarm can be cleared by manual confirmation or automatically after the risk returns to a safe range. The clearing record is saved simultaneously. Data storage and management: The system employs a dual storage model combining local storage and cloud backup, automatically storing patient data from admission to discharge. This includes raw data, preprocessed data, coupled risk analysis results, treatment instructions, alarm records, and physician operation records. Data storage is encrypted to prevent data leakage and ensure data security. The system supports reset operations; when a patient is discharged or transferred to another department, the physician can initiate a reset request via the terminal. The system will retain all historical data and then clear the current monitoring interface, preparing it for the next patient. It also supports data export, allowing data to be exported through a specified interface or the hospital intranet to meet the needs of clinical research, case analysis, or medical dispute tracing.

[0029] like Figure 2 As shown, the specific workflow of the hemodynamic dynamic monitoring and support system for critically ill ICU patients provided by this invention is as follows: After the patient is admitted to the hospital, the medical staff connect the sensor and acquisition tubing of the multi-parameter synchronous acquisition module to the patient's body, enter the patient's basic information, and the system collects the first data and, after standardization processing, uses it as the baseline parameters and stores it in the individual characteristic database. The multi-parameter synchronous acquisition module continuously acquires hemodynamic parameters and coagulation function indicators at a set frequency and transmits them to the data preprocessing module in real time for filtering, outlier removal and interpolation completion. The dynamic coupling analysis module calls the baseline data in the individual characteristic database, dynamically adjusts the early warning threshold, calculates the coupling risk value through the coupling analysis model, accurately determines the risk type and level, and simultaneously marks abnormal indicators. Based on the risk assessment results, the intelligent intervention and control module matches the optimal treatment plan from the intervention strategy library, generates control commands and transmits them to the treatment equipment, and simultaneously activates a timed feedback correction mechanism. The display alarm and data storage module shows monitoring data, risk status and treatment plan in real time, triggers corresponding audible and visual alarms according to the risk level, and stores the entire process data synchronously. Repeat the above steps to continuously monitor, analyze, intervene and provide feedback until the patient is discharged or transferred to another department, at which point the system completes the data retention and reset operation.

[0030] This invention enables dynamic monitoring, precise analysis, and intelligent intervention of hemodynamics and coagulation function in ICU critically ill patients through the collaborative work of multiple modules. It effectively reduces the risk of complications such as bleeding and thrombosis, provides a scientific basis for clinical treatment, and has significant clinical application value.

[0031] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A dynamic hemodynamic monitoring and support system for critically ill ICU patients, characterized in that, The system includes: Multi-parameter synchronous acquisition module: integrates hemodynamic acquisition unit and coagulation function acquisition unit, synchronously and frequently acquires patient hemodynamic parameters and periodically acquires coagulation function indicators with bedside detection equipment through adapter sensors, and transmits the raw data to data preprocessing module; Data preprocessing module: Cleans and standardizes the raw data through filtering, outlier identification and removal, and interpolation completion techniques, outputting continuous and accurate standardized data and transmitting it to the dynamic coupling analysis module; Dynamic Coupling Analysis Module: Based on an individual characteristic database, the module calculates coupling risk values ​​and determines risk types and levels by dynamically adjusting early warning thresholds and constructing a blood flow-coagulation correlation model. The analysis results are then transmitted to the intelligent intervention and control module. Intelligent intervention and control module: Based on the intervention strategy library, it matches differentiated treatment plans, generates and sends control commands for treatment equipment, and corrects the commands through a timed feedback and correction mechanism to ensure that the intervention effect is consistent with the changes in the patient's condition; Display alarm and data storage module: Real-time display of monitoring data, risk status and intervention instructions through interactive terminal, hierarchical audible and visual alarms based on coupled risk values, automatic storage of full-process data, and support for reset operation and data retention.

2. The hemodynamic monitoring and support system for critically ill ICU patients according to claim 1, characterized in that, The hemodynamic acquisition unit consists of an arterial pressure sensor, a central venous pressure sensor, a cardiac output monitoring sensor, a peripheral blood oxygen saturation sensor, and a body temperature sensor, and collects systolic blood pressure, diastolic blood pressure, mean arterial pressure, central venous pressure, cardiac output, stroke volume, peripheral vascular resistance, blood oxygen saturation, and core body temperature parameters in real time. The coagulation function collection unit is equipped with a thromboelastography instrument, a platelet function analyzer, a thrombin generation rate detector, and a fibrinogen concentration detector. It is connected to the patient's peripheral vein through a disposable sterile blood collection tubing to collect thrombin generation rate, fibrinogen level, platelet aggregation rate, and thromboelastography parameters, achieving synchronous acquisition of blood flow and coagulation indicators.

3. The hemodynamic monitoring and support system for critically ill ICU patients according to claim 1, characterized in that, The dynamic coupling analysis module has a built-in individual characteristic database and coupling analysis model. The individual characteristic database stores the patient's basic information as well as the baseline parameters of hemodynamics and coagulation function at the time of admission, providing a basis for personalized analysis; and dynamically adjusts the warning thresholds of each parameter according to the deviation between the patient's real-time parameters and the baseline data, and the time of disease progression. Simultaneously, an association model is constructed and trained based on clinical data. The weight coefficients and correction constants of each core association indicator are calibrated. Hemodynamic parameters and coagulation function indicators are selected as core association indicators. Differentiated weights are assigned to each indicator according to the patient's condition type. Coupled with the correction constant, the coupling risk value is calculated. The type of bleeding or thrombosis risk is determined by the positive or negative value of the coupling risk value. The risk level is divided by the absolute value. After the analysis is completed, the risk type, level and related parameter data are transmitted to the intelligent intervention and control module.

4. The hemodynamic monitoring and support system for critically ill ICU patients according to claim 3, characterized in that, The dynamic coupling analysis module dynamically adjusts the warning thresholds of each parameter based on the deviation between the patient's real-time parameters and baseline data, as well as the time of disease progression. The adjustment formula is as follows: ,in, It is the first The lower limit of the parameter warning threshold at any given time, i.e., the minimum safe threshold after dynamic adjustment. These are the baseline values ​​of the corresponding parameters when the patient was admitted to the hospital. This is the disease condition impact coefficient, with a value ranging from 0.1 to 0.3, which is automatically matched by the dynamic coupling analysis module based on the patient's disease type. It is the time decay coefficient. This refers to the progression of the disease.

5. The hemodynamic monitoring and support system for critically ill ICU patients according to claim 3, characterized in that, The dynamic coupling analysis module constructs an association model, the model formula of which is: ,in, This is a coupled risk value used to determine the type and level of risk. Negative values ​​indicate bleeding risk, while positive values ​​indicate thrombosis risk. The larger the absolute value, the higher the risk level. , These are the core hemodynamic parameters, among which Corresponding mean arterial pressure, Corresponding cardiac output; , , The core coagulation function indicators, among which Corresponding thrombin generation rate, Corresponding fibrinogen level, Corresponding platelet aggregation rate (PAgT); , It is the differential weighting of hemodynamic parameters. , , It is the differentiated weighting of coagulation function indicators. It is a correction constant, which is a preset system calibration value.

6. The hemodynamic monitoring and support system for critically ill ICU patients according to claim 3, characterized in that, The dynamic coupling analysis module is trained based on clinical data, calibrating the weight coefficients and correction constants of each core correlation indicator. The specific training steps are as follows: First, clinical data of ICU critically ill patients are collected, including corresponding hemodynamic parameters, coagulation function indicators, and clinical outcomes. After cleaning and removing outliers and standardizing the data, it is divided into training and validation sets according to the proportion. Using hemodynamic parameters and coagulation function indicators as input features and bleeding / thrombosis labels corresponding to clinical outcomes as output labels, a multivariate linear regression model framework is constructed. The weight coefficients and correction constants of each feature are iteratively optimized using the least squares method through the training set data to achieve the best match between the coupling risk value output by the model and the actual clinical risk status. Then, the model performance is evaluated using the validation set data, with accuracy, sensitivity, and specificity as core indicators. If the performance does not reach the preset threshold, the feature dimensions are readjusted or the algorithm parameters are optimized and the model is trained again. After training is complete, the model is embedded into the dynamic coupling analysis module.

7. The hemodynamic monitoring and support system for critically ill ICU patients according to claim 5, characterized in that, The It is the coupling risk value, when A value <0 indicates a bleeding risk; a negative value indicates that coagulation function-related indicators are dominant, or hemodynamic parameters (indicating bleeding tendency; the smaller the value, the higher the bleeding risk); when A value >0 indicates a risk of thrombosis. A positive value indicates that coagulation function-related indicators are dominant, or that hemodynamic parameters suggest a tendency to thrombosis. The higher the value, the higher the risk of thrombosis. No significant risk: when... =0 or close to 0, that is, when the absolute value is <5, it is determined that there is no significant risk of bleeding / thrombosis, and hemodynamics and coagulation function are in a relatively balanced state.

8. The hemodynamic monitoring and support system for critically ill ICU patients according to claim 1, characterized in that, The intelligent intervention and control module has a built-in intervention strategy library and correction mechanism. The intervention strategy library pre-stores differentiated intervention rules for different disease types, which can match the corresponding treatment plan according to different disease states and coupling risk results. Based on the risk judgment results transmitted by the dynamic coupling analysis module, the corresponding treatment plan is accurately matched from the strategy library to generate control instructions including fluid infusion rate and drug administration rate. These instructions are transmitted to the infusion pumps and drug administration devices in the ICU through the adapter bus. The module also has a timed feedback correction capability, which collects the adjusted parameters and recalculates the coupling risk value. At the same time, the execution status of the instructions is synchronously transmitted to the display alarm and data storage module.

9. The hemodynamic monitoring and support system for critically ill ICU patients according to claim 8, characterized in that, The intelligent intervention and control module collects the adjusted parameters and recalculates the coupling risk value. If the risk value returns to the safe range, the current treatment plan is maintained. If the risk value does not return to the safe range, the system will intelligently correct the instructions according to the magnitude of the risk change. If the risk value has decreased significantly compared to before the adjustment but has not reached the safe range, the current adjustment direction is maintained and the adjustment magnitude is optimized. If the risk value does not improve significantly or even increases, the system will immediately switch to the corresponding alternative treatment plan in the strategy library.

Images