Hemodialysis care unit extracorporeal circulator monitoring method and system and storage medium

Abstract

The invention relates to the technical field of data monitoring and analysis, and discloses a hemodialysis care unit extracorporeal circulator monitoring method and system and a storage medium. The method comprises the steps that vital signs and identity features of a patient are collected, a sign feature data set is obtained through fusion processing, and initial dialysis parameters are generated; analyzing the sign feature data set to obtain blood vessel state parameters; adjusting a circulation system based on the blood vessel state parameters and the initial dialysis parameters to obtain circulation control data; monitoring a dialysate path by using the circulation control data, and generating a liquid path operation parameter; the circulation control data and the liquid path operation parameters are integrated to analyze the system state, and equipment early warning indexes are formed; and evaluating equipment operation according to the early warning index and generating an automatic control instruction. According to the method, deep integration of the physical sign characteristic data of the patient and the equipment operation data is realized, a data-driven objective decision basis is provided for the dialysis process, and the safety and effectiveness of dialysis treatment are improved.

Description

Technical Field

[0001] The present application relates to the field of data monitoring and analysis, and in particular to a method, system and storage medium for monitoring an extracorporeal circulation device in a hemodialysis intensive care unit. Background Art

[0002] The existing monitoring methods for extracorporeal circulation devices in hemodialysis intensive care units mainly rely on the experience and judgment of medical staff and manual monitoring. The dialysis process requires medical staff to frequently check the patient's vital signs and the operating status of the equipment. Traditional monitoring methods usually use independent equipment to monitor the patient's vital signs and the operating parameters of the dialysis equipment respectively, lacking effective data fusion and comprehensive analysis mechanisms. Medical staff need to pay attention to multiple monitoring screens at the same time, and adjust the dialysis parameters based on experience to judge the correlation between various parameters. In addition, the existing extracorporeal circulation device monitoring system mainly relies on ultrasound or pressure waveform analysis to monitor the status of vascular access. These methods often require direct intervention on the patient, increasing the patient's discomfort and risk of infection.

[0003] However, the existing technology has obvious shortcomings: first, the lack of automated correlation analysis of patient identity characteristics and vital signs leads to data silos, making it difficult to form a complete patient feature data set; second, vascular access status assessment mainly relies on invasive testing or the experience and judgment of medical staff, and lacks an objective assessment method based on multi-parameter fusion; third, dialysis parameter adjustments are mostly manual operations, making it difficult to achieve dynamic optimization based on the patient's real-time status; fourth, the equipment abnormality warning mechanism is simple, mostly based on a single parameter threshold judgment, and lacks multi-dimensional cross-validation and trend analysis capabilities; finally, the equipment operation status assessment lacks statistical analysis in the time and space dimensions, making it difficult to achieve early warning and precise positioning of faults. Summary of the invention

[0004] The present application provides a method, system and storage medium for monitoring an extracorporeal circulation device in a hemodialysis intensive care unit, which is used to achieve deep integration of patient vital sign feature data and equipment operation data by establishing a monitoring method based on multi-source data fusion, provide a data-driven objective decision-making basis for the dialysis process, and improve the safety and effectiveness of dialysis treatment.

[0005] In a first aspect, the present application provides a method for monitoring an extracorporeal circulator in a hemodialysis intensive care unit, the method comprising: collecting vital signs and extracting identity features of a patient through a multi-parameter monitoring device, obtaining a patient vital sign feature data set through data fusion processing, and generating initial dialysis parameters based on the patient vital sign feature data set; performing vascular state analysis on the patient vital sign feature data set to generate vascular state parameters; dynamically adjusting the pressure value and blood flow value of the extracorporeal circulation system based on the vascular state parameters and the initial dialysis parameters, and obtaining circulation control data through parameter correction; based on the circulation control data, real-time monitoring of the water supply pressure, concentrate concentration and pipeline tightness of the dialysis fluid circuit, and obtaining fluid circuit operation parameters through data analysis; performing multi-dimensional analysis on the system operation status based on the circulation control data and the fluid circuit operation parameters, and obtaining equipment early warning indicators through threshold comparison; performing statistical analysis on the operation parameters of the extracorporeal circulator based on the equipment early warning indicators, and generating equipment operation evaluation data and automatic control instructions.

[0006] In a second aspect, the present application provides an extracorporeal circulator monitoring system for a hemodialysis intensive care unit, the extracorporeal circulator monitoring system for a hemodialysis intensive care unit comprising:

[0007] The acquisition module is used to collect the patient's vital signs and extract identity features through a multi-parameter monitoring device, obtain a patient's vital signs feature data set through data fusion processing, and generate initial dialysis parameters based on the patient's vital signs feature data set;

[0008] A detection module, used for performing vascular status analysis on the patient's physical sign feature data set to generate vascular status parameters;

[0009] A regulating module, used for dynamically regulating the pressure value and blood flow value of the extracorporeal circulation system according to the vascular state parameters and the initial dialysis parameters, and obtaining circulation control data through parameter correction;

[0010] A monitoring module, for real-time monitoring of the water supply pressure, concentrate concentration and pipeline tightness of the dialysate circuit based on the circulation control data, and obtaining the circuit operation parameters through data analysis;

[0011] An analysis module, used to perform multi-dimensional analysis on the system operation status according to the circulation control data and the fluid circuit operation parameters, and obtain equipment early warning indicators through threshold comparison;

[0012] The control module is used to perform statistical analysis on the operating parameters of the extracorporeal circulation device according to the equipment early warning indicators, and generate equipment operation evaluation data and automatic control instructions.

[0013] A third aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the above-mentioned method for monitoring an extracorporeal circulator in a hemodialysis intensive care unit.

[0014] In the technical solution provided by the present application, the patient's vital signs are collected and identity features are extracted through multi-parameter monitoring equipment, and the patient's biological information and physiological parameters are effectively integrated in combination with data fusion processing technology to form a comprehensive patient vital signs feature data set, avoiding the data island problem in traditional methods. The vascular state analysis based on the patient's vital signs feature data set innovatively applies a non-invasive assessment method, generates vascular state parameters through correlation analysis of physiological parameters such as blood pressure and heart rate, and improves the objectivity and accuracy of vascular state assessment. The mechanism of dynamically adjusting the extracorporeal circulation system based on vascular state parameters and initial dialysis parameters realizes precise control of pressure values ​​and blood flow values. The circulation control data obtained through parameter correction provides a strong guarantee for the stable operation of the system and reduces the risk of complications during dialysis. The application of circulation control data further supports the real-time monitoring of the water supply pressure, concentrate concentration and pipeline tightness of the dialysis fluid circuit, and the fluid circuit operation parameters obtained through data analysis improve the comprehensive monitoring of the equipment operation status. This solution innovatively combines circulation control data and fluid circuit operation parameters for multi-dimensional analysis. The equipment warning indicators generated by threshold comparison have high sensitivity and specificity, which significantly improves the early warning capability of abnormal situations. The equipment operation evaluation data and automatic control instructions generated based on the statistical analysis of the equipment warning indicators realize the self-optimization adjustment of the operation parameters of the extracorporeal circulator, reduce the workload of medical staff, and improve the safety and effectiveness of dialysis treatment. In particular, in the present invention, the application of artificial intelligence algorithms in data fusion, vascular status analysis and multi-dimensional anomaly detection enables the system to extract key features from massive monitoring data, identify potential abnormal patterns, and predict possible risks based on historical data and current status. The introduction of algorithm features significantly enhances the intelligence level of extracorporeal circulator monitoring, enables the system to have self-learning and self-adaptive capabilities, and can automatically adjust the monitoring strategy according to the individual differences of different patients, improve the accuracy and reliability of monitoring, and reduce operational errors by reducing human intervention. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0016] Figure 1A schematic diagram of an embodiment of a method for monitoring an extracorporeal circulation device in a hemodialysis intensive care unit in an embodiment of the present application;

[0017] Figure 2 This is a schematic diagram of an embodiment of an extracorporeal circulation monitoring system for a hemodialysis intensive care unit in an embodiment of the present application. DETAILED DESCRIPTION

[0018] Embodiments of the present application provide a method, system and storage medium for monitoring an extracorporeal circulation device in a ward of hemodialysis. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described here can be implemented in an order other than the content illustrated or described here. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0019] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 In the embodiment of the present application, an embodiment of the method for monitoring an extracorporeal circulation device in a hemodialysis intensive care unit includes:

[0020] Step S101, collecting vital signs and extracting identity features of the patient through a multi-parameter monitoring device, obtaining a patient vital sign feature data set through data fusion processing, and generating initial dialysis parameters based on the patient vital sign feature data set;

[0021] Step S102, performing vascular status analysis on the patient's physical sign feature data set to generate vascular status parameters;

[0022] Step S103, dynamically adjusting the pressure value and blood flow value of the extracorporeal circulation system according to the vascular state parameters and the initial dialysis parameters, and obtaining circulation control data through parameter correction;

[0023] Step S104: Based on the circulation control data, the water supply pressure, the concentrate concentration and the pipeline tightness of the dialysate circuit are monitored in real time, and the circuit operation parameters are obtained through data analysis;

[0024] Step S105: Perform multi-dimensional analysis on the system operation status according to the circulation control data and the fluid circuit operation parameters, and obtain the equipment early warning index by threshold comparison;

[0025] Step S106: Perform statistical analysis on the operating parameters of the extracorporeal circulation device based on the equipment early warning indicators to generate equipment operation evaluation data and automatic control instructions.

[0026] It is understandable that the execution subject of the present application may be an extracorporeal circulation monitoring system of a hemodialysis intensive care unit, or a terminal or a server, which is not specifically limited here. The present application embodiment is described by taking a server as the execution subject as an example.

[0027] Specifically, the patient's vital signs data, including heart rate, respiration, blood oxygen saturation, blood pressure and other indicators, are collected through multi-parameter monitoring equipment. Specifically, the heart rate parameter range is set to 50-130 times / min, the respiration parameter range is 10-30 times / min, the lower limit of the blood oxygen saturation parameter is 90%, the systolic pressure parameter range is 85-240mmHg, and the diastolic pressure parameter range is 50-120mmHg. At the same time, the patient's facial features and fingerprint features are collected for identity recognition. These data are fused and processed to generate a patient's vital signs feature data set. On this basis, combined with the patient's weight data, the weight difference is calculated to generate the initial dialysis parameters.

[0028] Blood pressure values ​​and heart rate values ​​are extracted from the patient's physical sign feature data set, and the blood pressure-heart rate multiplication factor is calculated by analyzing the relationship between the two. The difference between systolic and diastolic pressures is multiplied by the heart rate and then divided by the standard coefficient 100 to obtain the basic vascular assessment data, which reflects the comprehensive state of vascular wall elasticity and cardiac load. For example, when the patient's systolic blood pressure is 120 mmHg, the diastolic blood pressure is 80 mmHg, and the heart rate is 75 times / minute, the calculated multiplication factor is (120-80)×75÷100=30, which is within the normal range (25-35), indicating that the vascular state is good. Next, the pulse pressure difference and blood pressure fluctuation trend are extracted from the patient's physical sign feature data set. By analyzing the amplitude and frequency of blood pressure changes in a continuous time period, an evaluation index reflecting the vascular resistance state is established to form blood flow state prediction data. It can predict the working state of the vascular access. At the same time, the body temperature value in the physical sign feature data set is associated with the blood oxygen saturation data, and a temperature-blood oxygen relationship map is constructed to generate temperature-related data. This data reflects the efficiency of blood circulation by recording the correspondence between changes in body temperature and blood oxygen saturation. The resistance index in the blood flow status prediction data is divided into time periods and counted, the fluctuation range in different time periods is calculated and the stability score is evaluated. When the fluctuation exceeds the preset threshold (such as the standard deviation is greater than 15%), it is marked to form blood flow stability data. This marking helps to identify potential vascular access abnormalities. Subsequently, the temperature-related data and blood flow stability data are comprehensively analyzed, and the vascular patency coefficient is determined by calculating the correlation coefficient of the two sets of data to generate vascular access assessment data. The vascular access assessment data and the vascular basic assessment data are weighted (such as the access assessment data weight 0.6, the basic assessment data weight 0.4), and the results are normalized to the range of 0-100 to form vascular status parameters.

[0029] The extracorporeal circulation system is adjusted according to the vascular state parameters and the initial dialysis parameters. The two sets of parameters are combined to generate the initial adjustment parameters of the system, from which the pressure control parameters are extracted: arterial pressure threshold -200-0mmHg, venous pressure threshold 50-200mmHg, transmembrane pressure threshold 0-400mmHg. The real-time pressure value is collected by the pressure sensor and compared with the pressure control parameters to obtain the pressure deviation data. According to the pressure deviation data, the blood flow velocity is adjusted according to the speed gradient of 50ml / min to generate blood flow regulation data. The pressure deviation data is cross-validated with the blood flow regulation data, and the parameter points that exceed the safety range are marked to generate parameter correction data and obtain circulation control data.

[0030] When the dialysis fluid circuit is monitored in real time, the pressure threshold and blood flow standard value are extracted from the circulation control data to generate the fluid circuit monitoring benchmark data. The water supply pipeline pressure is sampled by the pressure sensor, and the pressure fluctuation analysis is performed with the fluid circuit monitoring benchmark data to obtain the water supply pressure data. The conductivity of the dialysis fluid concentrate is detected, and the concentrate concentration data is generated by comparing the conductivity value with the standard ratio. The pressure sensor array is used to detect the air pressure of each node in the pipeline, and the detection values ​​are arranged in time series to generate pipeline tightness data. The three sets of data are correlated and compared, and the abnormal change points are marked to obtain the abnormal data of the fluid circuit, and then the fluid circuit operation parameters are generated. The system operation status is analyzed based on the circulation control data and the fluid circuit operation parameters. The pressure values, blood flow values, water supply pressure, concentrate concentration, and pipeline tightness data in the two sets of data are integrated to obtain the system operation data. The pressure fluctuation value, blood flow stability value, concentration change value, and tightness detection value are extracted from it to generate parameter monitoring data. The parameter monitoring data is segmented according to the time series, and the data fluctuation amplitude in each time period is compared with the standard threshold to generate fluctuation analysis data. Extract features from abnormal points in the fluctuation analysis data, calculate parameter change trends, and obtain trend prediction data. Compare the change rate in the trend prediction data with the safety threshold, mark the parameter points that exceed the threshold, generate risk parameter data, and obtain equipment warning indicators based on this.

[0031] Generate evaluation data and control instructions based on the equipment early warning indicators. Extract abnormal parameters from the early warning indicators, analyze their changing patterns, and obtain parameter change characteristic data. Classify the parameter change characteristic data according to the system functional modules, calculate the parameter correlation, and generate system performance evaluation data. Statistically analyze the fluctuation range of each operating parameter to obtain parameter distribution data. Compare the discrete points in the parameter distribution data with the standard operating parameters, mark the operating parameters whose deviation values ​​exceed the threshold, and generate abnormal operation data. Perform multi-dimensional statistical analysis based on the time distribution and spatial distribution characteristics of the abnormal operation data to obtain the equipment operation evaluation data. Generate automatic control instructions for the extracorporeal circulator by performing parameter optimization analysis on the equipment operation evaluation data.

[0032] For example, when the patient's heart rate is detected to be 45 beats / minute, the system immediately marks it as an abnormal value; when the blood flow velocity is detected by the ultrasound probe to be 220ml / min, it is determined as a blood flow abnormality point; when the pressure sensor detects that the arterial pressure is -250mmHg, the pressure abnormality alarm is triggered. After these abnormal data are processed, corresponding adjustment instructions are generated, such as reducing the blood flow velocity to 180ml / min, adjusting the arterial pressure to -180mmHg, etc., so as to maintain the stable operation of the extracorporeal circulation system.

[0033] In the embodiment of the present application, the patient's vital signs are collected and identity features are extracted through multi-parameter monitoring equipment, and the patient's biological information and physiological parameters are effectively integrated in combination with data fusion processing technology to form a comprehensive patient vital signs feature data set, avoiding the data island problem in traditional methods. The vascular state analysis based on the patient's vital signs feature data set innovatively applies a non-invasive assessment method, generates vascular state parameters through correlation analysis of physiological parameters such as blood pressure and heart rate, and improves the objectivity and accuracy of vascular state assessment. The mechanism of dynamically adjusting the extracorporeal circulation system based on vascular state parameters and initial dialysis parameters realizes precise control of pressure values ​​and blood flow values. The circulation control data obtained through parameter correction provides a strong guarantee for the stable operation of the system and reduces the risk of complications during dialysis. The application of circulation control data further supports the real-time monitoring of the water supply pressure, concentrate concentration and pipeline tightness of the dialysis fluid circuit, and the fluid circuit operation parameters obtained through data analysis improve the comprehensive monitoring of the equipment operation status. This solution innovatively combines circulation control data and fluid circuit operation parameters for multi-dimensional analysis. The equipment warning indicators generated by threshold comparison have high sensitivity and specificity, which significantly improves the early warning capability of abnormal situations. The equipment operation evaluation data and automatic control instructions generated based on the statistical analysis of the equipment warning indicators realize the self-optimization adjustment of the operation parameters of the extracorporeal circulator, reduce the workload of medical staff, and improve the safety and effectiveness of dialysis treatment. In particular, in the present invention, the application of artificial intelligence algorithms in data fusion, vascular status analysis and multi-dimensional anomaly detection enables the system to extract key features from massive monitoring data, identify potential abnormal patterns, and predict possible risks based on historical data and current status. The introduction of algorithm features significantly enhances the intelligence level of extracorporeal circulator monitoring, enables the system to have self-learning and self-adaptive capabilities, and can automatically adjust the monitoring strategy according to the individual differences of different patients, improve the accuracy and reliability of monitoring, and reduce operational errors by reducing human intervention.

[0034] In a specific embodiment, the process of executing step S101 may specifically include the following steps:

[0035] (1) Collecting facial features and fingerprint features of the patient through a biometric collection device to generate biometric data;

[0036] (2) Sampling the patient's heart rate, respiration, blood oxygen saturation, and blood pressure values ​​to generate physiological parameter data;

[0037] (3) Compare and verify the facial features and fingerprint features in the biometric data to obtain the identity authentication result;

[0038] (4) comparing each parameter value in the physiological parameter data with a preset parameter range, wherein the heart rate parameter range is 50-130 times / min, the breathing parameter range is 10-30 times / min, the lower limit of the blood oxygen saturation parameter is 90%, the systolic blood pressure parameter range is 85-240 mmHg, and the diastolic blood pressure parameter range is 50-120 mmHg;

[0039] (5) Integrate the identity verification results and physiological parameter data to form a patient vital sign feature data set;

[0040] (6) The patient's weight data is collected through a weight sensor device, and the weight difference is calculated based on the physiological parameter data in the patient's physical sign characteristic data set, and the initial dialysis parameters are generated according to the weight difference.

[0041] Specifically, the patient is identified through a biometric acquisition device. The biometric acquisition device includes a high-definition camera and a fingerprint scanner. The high-definition camera collects the patient's facial feature point data, including the position information of key points such as the corners of the eyes, the corners of the mouth, and the wings of the nose. The fingerprint scanner collects fingerprint pattern data, including the direction of the fingerprint ridges, the distribution of feature points, etc. These data are digitized to generate biometric data in a standard format. At the same time, the multi-parameter physiological monitoring equipment collects the patient's vital signs. The heart rate data is collected through the electrocardiogram monitor with a sampling frequency of 250Hz, and the average heart rate value is calculated every 1 minute; the respiratory rate is collected through the respiratory monitor with a sampling frequency of 100Hz, and the average respiratory value is calculated every 1 minute; the blood oxygen saturation is collected through the blood oxygen probe with a sampling frequency of 60Hz, and the average blood oxygen value is calculated every 1 minute; the systolic and diastolic blood pressures are collected through the sphygmomanometer and measured every 5 minutes. These data are filtered and denoised to generate physiological parameter data.

[0042] The authentication process of biometric data includes two steps: feature extraction and matching. The feature extraction step converts facial feature point data and fingerprint pattern data into feature vectors, and the matching step calculates the similarity between the feature vectors and the samples in the database to generate the authentication result. The abnormality detection process of physiological parameter data includes parameter range check and trend analysis. The parameter range check compares each physiological indicator with the preset safety range: heart rate exceeds 50-130 beats / min, respiration exceeds 10-30 times / min, blood oxygen saturation is less than 90%, systolic blood pressure exceeds 85-240mmHg, and diastolic blood pressure exceeds 50-120mmHg, which are marked as abnormal. Trend analysis calculates the rate of change of parameters, and it is also marked as abnormal when the rate of change exceeds the threshold.

[0043] For the calculation of weight difference, the following formula is used:

[0044]

[0045] Among them, ΔW is the weight difference, W c is the current weight, W b is the basic body weight, ω1 and ω2 are weight coefficients, V i is the value of the ith physiological parameter, α i is the corresponding influencing factor, and n is the total number of physiological parameters.

[0046] Initial dialysis parameters were generated using the following formula:

[0047]

[0048] Among them, P init is the initial dialysis parameter, β1, β2, β3 are adjustment coefficients, H j is the jth hemodynamic parameter, γ j is the corresponding conversion coefficient, m is the total number of hemodynamic parameters, Q k is the kth physiological state index, λ k is the corresponding correction coefficient, and p is the total number of physiological status indicators.

[0049] For example, when a patient undergoes hemodialysis, facial images and fingerprint data are collected through a biometric collection device. Facial feature extraction obtains 150 feature point coordinates, and fingerprint feature extraction obtains 80 feature point positions. These feature data are matched with samples in the database to calculate the similarity score. At the same time, the monitoring equipment collects the patient's heart rate of 75 times / min, breathing 20 times / min, blood oxygen saturation 95%, systolic blood pressure 130mmHg, and diastolic blood pressure 85mmHg. After data processing, these data are compared with the safety range and are all within the normal range. The patient's current weight is measured by an electronic scale to be 72kg and the basic weight is 70kg. The weight difference is calculated in combination with the physiological parameter data, and then the initial dialysis parameters suitable for the patient are generated.

[0050] In a specific embodiment, the process of executing step S102 may specifically include the following steps:

[0051] (1) extracting the blood pressure values ​​and heart rate values ​​in the patient's physical sign feature data set, analyzing the ratio relationship between blood pressure and heart rate, calculating the blood pressure-heart rate multiplication factor, and generating vascular basic assessment data;

[0052] (2) extracting pulse pressure difference and blood pressure fluctuation trend from the patient's physical sign feature data set, establishing a blood flow resistance evaluation index, and generating blood flow state prediction data;

[0053] (3) performing correlation analysis on the body temperature value in the patient's physical sign feature data set and the blood oxygen saturation change in the physiological parameter data, constructing a relationship map between temperature and blood oxygen, and generating temperature correlation data;

[0054] (4) performing segmented statistics on the resistance index in the blood flow state prediction data, calculating the fluctuation range to obtain a stability score, marking the fluctuation amplitude that exceeds a preset threshold, and generating blood flow stability data;

[0055] (5) performing parameter correlation analysis on the temperature-related data and the blood flow stability data, calculating the vascular patency coefficient, and generating vascular access assessment data;

[0056] (6) Comprehensively weighting the vascular pathway assessment data and the vascular basic assessment data, and generating vascular status parameters by data normalization.

[0057] Specifically, blood pressure values ​​and heart rate values ​​are extracted from the patient's physical sign feature data set. These values ​​are collected by multi-parameter monitoring equipment and stored in the data set. After extraction, the ratio relationship between blood pressure and heart rate is analyzed, mainly calculating the difference between systolic pressure and diastolic pressure, and then multiplying it with the heart rate to form a blood pressure-heart rate multiplication factor. This factor reflects the comprehensive state of vascular wall elasticity and cardiac load. The calculation process multiplies the difference by the heart rate and then divides it by the standard coefficient to obtain the basic vascular assessment data. For example, when the patient's systolic blood pressure is 135 mmHg, the diastolic blood pressure is 85 mmHg, and the heart rate is 78 beats / minute, the calculated multiplication factor will reflect the basic information of the current vascular state.

[0058] The pulse pressure difference and blood pressure fluctuation trend are extracted from the patient's physical sign feature data set. The pulse pressure difference refers to the difference between systolic pressure and diastolic pressure, and the blood pressure fluctuation trend refers to the change of blood pressure measurement values ​​at multiple consecutive time points. By analyzing these data, a blood flow resistance evaluation index is established. This index analyzes the time series changes of the pulse pressure difference and combines the change rate of systolic pressure and diastolic pressure to generate blood flow state prediction data reflecting the state of vascular resistance. The body temperature values ​​in the patient's physical sign feature data set and the changes in blood oxygen saturation in the physiological parameter data are correlated and analyzed. During the analysis process, the body temperature values ​​at consecutive time points are paired with the blood oxygen saturation data at the corresponding time points, and the correlation between the two sets of data is analyzed to construct a two-dimensional map reflecting the relationship between temperature and blood oxygen. The horizontal axis of the map represents the change in body temperature, and the vertical axis represents the change in blood oxygen saturation. Each point on the map represents the state at a specific time. Through this mapping relationship, temperature-related data is generated, which reflects the degree of influence of body temperature changes on blood oxygen saturation.

[0059] The resistance index in the blood flow state prediction data is divided into time periods, usually with 10 minutes as a time unit, and the maximum, minimum and average values ​​of the resistance index in each time period are calculated to obtain the fluctuation range. Based on the size of the fluctuation range, a corresponding stability score is given. The scoring standard is usually 1-10 points, and the smaller the fluctuation, the higher the score. When the fluctuation amplitude in a certain time period exceeds the preset threshold (such as the standard deviation exceeds 15%), the system will specially mark the interval. These marking information together with the corresponding stability score constitute the blood flow stability data. Parameter correlation analysis is performed on temperature-related data and blood flow stability data. During the analysis process, the temperature change trend in the temperature-related data is time-aligned with the fluctuation in the blood flow stability data, and the correlation coefficient between the two sets of data is calculated. The coefficient reflects the strength of the correlation between temperature change and blood flow stability. The vascular patency coefficient is calculated by combining the correlation coefficient with the weight factor. The closer the coefficient is to 1, the better the vascular patency state is, and the vascular access assessment data is generated accordingly.

[0060] The vascular access assessment data and the vascular basic assessment data are comprehensively weighted. During the weighting process, the weight of the vascular access assessment data is usually set to 0.6, and the weight of the vascular basic assessment data is set to 0.4. After the weighted sum of the two parts, the result is mapped to the range of 0-100 through the normalization function to form the final vascular status parameter. This parameter intuitively reflects the real-time status of the vascular access and provides a scientific basis for adjusting the dialysis parameters. In actual applications, when the vascular status parameter is continuously lower than 60 points, the system will recommend that medical staff check the patient's vascular access status to prevent complications during dialysis. The entire analysis process avoids direct intervention of traditional methods on patients through multi-dimensional processing of patient data, thereby improving the safety and effectiveness of the dialysis process.

[0061] In a specific embodiment, the process of executing step S103 may specifically include the following steps:

[0062] (1) Merging the vascular status parameters with the initial dialysis parameters to generate the initial adjustment parameters of the system;

[0063] (2) extracting the arterial pressure threshold -200-0 mmHg, the venous pressure threshold 50-200 mmHg, and the transmembrane pressure threshold 0-400 mmHg from the initial adjustment parameters of the system to generate pressure control parameters;

[0064] (3) collecting the real-time pressure value in the extracorporeal circulation system through a pressure sensor, comparing it with the pressure control parameter, and obtaining pressure deviation data;

[0065] (4) Based on the pressure deviation data, the blood flow velocity of the extracorporeal circulation system is divided into intervals, and the velocity gradient is adjusted according to 50 ml / min to generate blood flow adjustment data;

[0066] (5) cross-validating the pressure deviation data with the blood flow regulation data, marking the parameter points that exceed the safety range, and generating parameter correction data;

[0067] (6) Generate loop control data based on the correspondence between the parameter correction data and the system initial adjustment parameters.

[0068] Specifically, the vascular status parameters and initial dialysis parameters are merged. The vascular status parameters include the vascular access function index, thrombosis risk index and stenosis index, and the initial dialysis parameters include the target ultrafiltration volume, dialysis time and blood flow velocity target value. The data merging process uses the data structure reorganization method to align the two sets of parameters according to the timestamp to form a unified data structure and generate the initial adjustment parameters of the system.

[0069] When extracting the pressure control parameters from the initial adjustment parameters of the system, three key pressure parameters are set according to the working characteristics of the dialysis machine: arterial pressure, venous pressure and transmembrane pressure. The arterial pressure threshold is set in the range of -200-0mmHg, because negative pressure is required for blood extraction at the arterial end; the venous pressure threshold is set in the range of 50-200mmHg, which is the positive pressure range required for blood return; the transmembrane pressure threshold is set in the range of 0-400mmHg, which is the pressure range to ensure the adequacy of dialysis. The setting of these thresholds is based on clinical experience and the technical specifications of the dialysis machine. The layout of the pressure sensor adopts a multi-point measurement strategy, and pressure sensors are installed at the arterial end, venous end and both ends of the filter, with a sampling frequency of 10Hz. The comparison of the real-time pressure value and the pressure control parameter adopts the sliding window method, and the window length is 30 seconds. In each time window, the mean, standard deviation and change trend of the pressure value are calculated. When any pressure parameter exceeds the threshold range or abnormal fluctuation occurs, the pressure deviation value is recorded, and the occurrence time and duration are marked to form the pressure deviation data.

[0070] When adjusting the blood flow velocity based on the pressure deviation data, the blood flow velocity range is divided into multiple intervals according to the gradient of 50ml / min. Starting from the lowest 50ml / min, set the velocity gradient points to 50, 100, 150, 200ml / min in sequence. When the pressure abnormality is detected, select the corresponding speed adjustment interval according to the direction and size of the pressure deviation. Strictly follow the step value of 50ml / min when adjusting up or down to avoid sudden changes in blood flow velocity. Monitor the pressure changes after each adjustment, record the parameter changes before and after the adjustment, and generate blood flow regulation data. The cross-validation of the pressure deviation data and the blood flow regulation data adopts the time series correlation analysis method. Align the two sets of data on the time axis to analyze the correspondence between the pressure change and the blood flow regulation. Set the safety range verification rule: when the pressure value exceeds the threshold range and cannot return to the safety range after blood flow adjustment, mark the parameter point as an abnormal point. Feature extraction is performed on these abnormal points, and the abnormal type, occurrence time, duration and severity are recorded to generate parameter correction data.

[0071] The parameter correction data is matched and analyzed with the initial adjustment parameters of the system. The correction coefficient is calculated by comparing the difference between the abnormal parameters and the initial setting values. The correction coefficient is applied to the initial parameters to obtain a new set of control parameters. These parameters are optimized through multiple rounds of iterations to generate cyclic control data.

[0072] For example, during a hemodialysis session, the vascular status parameters showed that the patient's vascular access function index was 85 points (out of 100 points), the thrombosis risk index was at a moderate level, and the stenosis index was mild. The initial dialysis parameters set the target ultrafiltration volume to 2.5L, the dialysis time to 4 hours, and the target blood flow rate to 200ml / min. These two sets of parameters were combined to generate the initial adjustment parameters of the system. After the start of dialysis, the arterial pressure sensor detected that the pressure value fluctuated between -150mmHg and -180mmHg, the venous pressure was between 140-160mmHg, and the transmembrane pressure was between 250-300mmHg. When the arterial pressure suddenly dropped to -220mmHg, the pressure abnormality alarm was triggered. Based on the pressure deviation data, the blood flow rate was reduced from 200ml / min to 150ml / min, and the arterial pressure was observed to rise back to -170mmHg. After cross-validation, it was confirmed that the adjustment measure was effective, but because the pressure was still slightly lower than the ideal value, parameter optimization was continued to generate stable circulation control data.

[0073] In a specific embodiment, the process of executing step S104 may specifically include the following steps:

[0074] (1) Extracting pressure thresholds and blood flow standard values ​​from circulation control data to generate fluid circuit monitoring benchmark data;

[0075] (2) Sampling the water supply pipeline pressure through a pressure sensor, performing pressure fluctuation analysis based on the liquid line monitoring benchmark data, and obtaining water supply pressure data;

[0076] (3) Conducting conductivity testing on the dialysate concentrate and generating concentrate concentration data by comparing the conductivity value with the standard ratio;

[0077] (4) Using a pressure sensor array to detect air pressure at each node of the pipeline, arranging the detection values ​​in time series to generate pipeline tightness data;

[0078] (5) Correlate and compare the water supply pressure data, concentrate concentration data, and pipeline tightness data, mark abnormal change points, and obtain abnormal liquid circuit data;

[0079] (6) Generate fluid circuit operation parameters based on the comparison results between the fluid circuit abnormal data and the fluid circuit monitoring benchmark data.

[0080] Specifically, the pressure threshold and blood flow standard value are extracted from the circulation control data, where the pressure threshold includes the normal working range of the water supply pressure, the standard range of the conductivity and the allowable deviation of the pipeline air tightness, and the blood flow standard value reflects the dialysate flow requirement required during the dialysis process. These data are reorganized through the data structure and organized into a standard format according to the time series to form the fluid circuit monitoring benchmark data.

[0081] The water supply pipeline pressure is monitored using a high-precision pressure sensor with a sampling frequency of 100Hz. The sensors are installed at key nodes of the water supply pipeline, including the water inlet, the outlet of the pretreatment unit, and the inlet of the dialyzer. The sampled pressure data is filtered to remove noise with a filter frequency band of 0.1-10Hz. The filtered data is segmented into 1-minute units, and the average pressure value and pressure fluctuation range in each segment are calculated and compared with the pressure threshold in the liquid circuit monitoring benchmark data to generate water supply pressure data.

[0082] The conductivity of the dialysate concentrate is calculated using the following formula:

[0083]

[0084] Among them, C dia is the dialysate concentration, R i is the conductivity reading of the ith electrolyte, E i is the corresponding electrolyte concentration conversion coefficient, K i is the temperature compensation coefficient, T j is the temperature value of the jth temperature measurement point, S j is the temperature sensitivity coefficient, μ is the correction coefficient, n is the number of electrolyte types, and m is the number of temperature measurement points.

[0085] The pipeline tightness detection uses a pressure sensor array, and pressure sensors are installed at multiple key nodes of the pipeline. Each sensor collects 10 data points per second to record the changes in air pressure values. The collected data is arranged in chronological order to form a pressure change curve. At the same time, the influencing factors such as ambient temperature and humidity are recorded, and the data is temperature compensated and humidity corrected to generate pipeline tightness data. The generation of abnormal liquid circuit data adopts a multi-dimensional anomaly detection method. The water supply pressure data, concentrate concentration data and pipeline tightness data are aligned on the time axis and analyzed by sliding windows. The window length is set to 5 minutes, and it slides every 30 seconds. In each window, the correlation coefficient of the three sets of data is calculated, and the time point of abnormal correlation is marked. At the same time, the data is trended. When a parameter suddenly changes or continuously deviates from the normal range, the time point is marked as an abnormal point.

[0086] The fluid circuit operation parameters are generated using the following formula:

[0087]

[0088] Among them, P dia is the fluid circuit operation parameter, F x is the weight coefficient of the xth abnormal index, M x is the corresponding abnormality value, D y is the deviation coefficient of the yth reference parameter, N y is the corresponding reference value, G z is the zth correction factor, O z is the corresponding correction value, l is the number of abnormal indicators, k is the number of benchmark parameters, and h is the number of correction factors.

[0089] For example: In a hemodialysis treatment, the standard range of water supply pressure extracted from the circulation control data is 200-400mmHg, and the dialysate flow requirement is 500ml / min. After starting monitoring, the water supply line pressure sensor detected that the pressure value fluctuated within the normal range, but there was a brief pressure drop. At the same time, the conductivity test showed that the dialysate concentration was slightly higher than the standard value. After calculation, it was found that the temperature compensation coefficient needed to be adjusted. The pipeline tightness test found that the pressure value at a certain interface fluctuated periodically. Through correlation analysis, it was determined that this was time-series correlated with the fluctuation of water supply pressure. Combining these abnormal data and combining them with the baseline parameters for calculation, new fluid circuit operation parameters are generated to guide the next stage of dialysis fluid circuit control.

[0090] In a specific embodiment, the process of executing step S105 may specifically include the following steps:

[0091] (1) Integrate the pressure value and blood flow value in the circulation control data with the water supply pressure, concentrate concentration, and pipeline tightness data in the liquid circuit operation parameters to obtain system operation data;

[0092] (2) Extracting pressure fluctuation values, blood flow stability values, concentration change values, and airtightness detection values ​​from system operation data to generate parameter monitoring data;

[0093] (3) The parameter monitoring data is segmented according to the time series, and the data fluctuation amplitude in each time period is compared with the standard threshold to generate fluctuation analysis data;

[0094] (4) Extract features of abnormal points in the fluctuation analysis data through data analysis, calculate the change trend of various parameters, and obtain trend prediction data;

[0095] (5) Compare the rate of change in the trend prediction data with the safety threshold, mark the parameter points that exceed the threshold, and generate risk parameter data;

[0096] (6) Generate equipment warning indicators based on the time series distribution and change characteristics of risk parameter data.

[0097] Specifically, the operation data of multiple subsystems are integrated and analyzed. The circulation control data and the liquid circuit operation parameters are fused, and the data structure is unified. Specifically, the pressure values ​​in the circulation control data include arterial pressure, venous pressure and transmembrane pressure, and the blood flow values ​​include real-time blood flow velocity and blood flow fluctuation range; the liquid circuit operation parameters include water supply pressure, concentrate concentration and pipeline tightness data. These data are aligned according to a unified timestamp to form a standardized data structure to generate system operation data. When extracting key parameters from the system operation data, the characteristic value extraction method is adopted. The pressure fluctuation value is obtained by calculating the standard deviation and coefficient of variation of the pressure data; the blood flow stability value is obtained by analyzing the fluctuation amplitude and frequency of the blood flow velocity; the concentration change value is obtained by calculating the first-order difference and change rate of the concentration data; the tightness detection value is obtained by analyzing the drift degree of the pipeline air pressure. These characteristic values ​​are sorted in a unified format to generate parameter monitoring data.

[0098] The time series analysis of parameter monitoring data adopts a segmented processing method. Every 5 minutes is regarded as a time period, and the data in each time period is statistically analyzed. The maximum value, minimum value, average value and standard deviation of each parameter in the time period are calculated and compared with the preset standard thresholds. The standard thresholds include arterial pressure range -200-0mmHg, venous pressure range 50-200mmHg, transmembrane pressure range 0-400mmHg, blood flow velocity fluctuation range ±20ml / min, concentration deviation range ±5%, air pressure change range ±10mmHg, etc. The comparison results form fluctuation analysis data.

[0099] The feature extraction of fluctuation analysis data adopts multi-dimensional analysis method. The data is detected for abnormal points by combining the moving average method and threshold detection. When a parameter value exceeds the range of 3 times the standard deviation of the moving average, it is marked as a potential abnormal point. Then these abnormal points are clustered to identify abnormal patterns and types. At the same time, trend analysis is performed on various parameters, and the linear regression method is used to calculate the change trend, predict the future change direction of the parameters, and form trend prediction data. The security assessment of trend prediction data adopts a graded warning method. Different levels of safety thresholds are set according to the change rate of each parameter. For example, the pressure change rate exceeds 10mmHg per minute for the first warning, and exceeds 20mmHg for the second warning; the blood flow velocity change rate exceeds 30ml / min per minute for the first warning, and exceeds 50ml / min for the second warning. When the parameter change rate exceeds the safety threshold of the corresponding level, the parameter point is marked as a risk point, and the risk level and duration are recorded to generate risk parameter data.

[0100] The generation of equipment early warning indicators adopts a comprehensive evaluation method. The risk parameter data is analyzed in time series to calculate the frequency and distribution characteristics of risk events. At the same time, the correlation between risk parameters is analyzed, such as the correlation between pressure abnormalities and blood flow fluctuations, and the correlation between concentration changes and pipeline tightness. Based on these analysis results, the priority and urgency of the warning are determined, and the equipment early warning indicators are generated.

[0101] For example, during a hemodialysis treatment, multiple operating data of the extracorporeal circulation system were integrated. The circulation control data showed that the arterial pressure value fluctuated around -150 mmHg, the venous pressure was maintained at 130 mmHg, and the blood flow rate was set at 200 ml / min; the liquid circuit operation parameters showed that the water supply pressure was normal, the concentrate concentration was stable, and the pipeline was well sealed. After time alignment, these data formed complete system operation data. During the parameter monitoring process, it was found that the fluctuation amplitude of the arterial pressure gradually increased, while the blood flow rate fluctuated slightly. Through time series analysis, it was found that the abnormalities of these two parameters were time-related. Further analysis found that the trend of arterial pressure changes pointed to a negative increase, and it was predicted that if no intervention was performed, the pressure value would exceed the safe range within 10 minutes. Based on these analysis results, a secondary warning indicator was generated to promptly remind medical staff to pay attention to the functional status of the vascular access.

[0102] In a specific embodiment, the process of executing step S106 may specifically include the following steps:

[0103] (1) Extract abnormal parameters from equipment warning indicators, perform time series analysis on parameter change patterns, and obtain parameter change characteristic data;

[0104] (2) Classify the parameter change characteristic data according to the system functional modules, calculate the parameter correlation, and generate system performance evaluation data;

[0105] (3) Based on the system performance evaluation data, the fluctuation range of each operating parameter is statistically analyzed to obtain parameter distribution data;

[0106] (4) comparing the discrete points in the parameter distribution data with the standard operating parameters, marking the operating parameters whose deviation values ​​exceed the threshold, and generating abnormal operating data;

[0107] (5) Perform multi-dimensional statistical analysis based on the temporal and spatial distribution characteristics of abnormal operation data to obtain equipment operation evaluation data;

[0108] (6) Generate automatic control instructions for the extracorporeal circulation device by performing parameter optimization analysis on the equipment operation evaluation data.

[0109] Specifically, parameters that exceed the threshold range are screened out from the equipment warning indicators, including abnormal pressure, abnormal blood flow, abnormal concentration, and abnormal tightness. These abnormal parameters are arranged in chronological order, and the sliding window method is used for time series analysis. The window length is set to 10 minutes, and the sliding step is 1 minute. In each window, the frequency, duration, and severity of abnormal parameters are calculated to form parameter change characteristic data. When the parameter change characteristic data is classified according to the system functional modules, the parameters are divided into three categories: blood circulation module, dialysate module, and control module. The blood circulation module includes parameters such as arterial pressure, venous pressure, transmembrane pressure, and blood flow velocity; the dialysate module includes parameters such as water supply pressure, concentrate concentration, and pipeline tightness; the control module includes various control instructions and feedback signals. The parameters in each module are analyzed for correlation, and the correlation coefficient method is used to calculate the correlation strength between the parameters. At the same time, the temporal dependency between the parameters is considered to generate system performance evaluation data.

[0110] The fluctuation range statistics of each parameter in the system performance evaluation data adopts the distribution feature analysis method. The statistical features such as mean, standard deviation, skewness and kurtosis are calculated for each parameter. At the same time, the fluctuation frequency characteristics of the parameters are analyzed, including periodic fluctuations, random fluctuations and trend changes. These statistical features and fluctuation features are combined into parameter distribution data to evaluate the stability of system operation. The identification and abnormal marking process of discrete points adopts the multi-threshold judgment method. Each data point in the parameter distribution data is compared with the standard operating parameters, which include the preset safety range and the dynamically adjusted target value. When the parameter deviation exceeds the warning threshold, primary marking is performed; when it exceeds the alarm threshold, secondary marking is performed; when it exceeds the danger threshold, advanced marking is performed. These marking information, together with the specific deviation value, form the operation abnormality data.

[0111] The multi-dimensional statistical analysis of operation abnormality data adopts the time-space correlation analysis method. In the time dimension, the time series characteristics of abnormal events are analyzed, including the occurrence time, duration, repetition cycle, etc.; in the space dimension, the distribution characteristics of abnormal events are analyzed, including the spatial correlation of abnormal parameters, propagation path, etc. Through this multi-dimensional analysis, more comprehensive equipment operation evaluation data can be obtained. The generation of automatic control instructions adopts the optimization decision-making method. The equipment operation evaluation data is classified and processed to identify the abnormal situations that need to be handled first. Then, according to the preset control strategy and parameter optimization rules, the corresponding control instructions are generated. Control instructions include parameter adjustment instructions, working mode switching instructions, and alarm processing instructions.

[0112] For example, during a hemodialysis treatment, the equipment warning indicator showed abnormal fluctuations in arterial pressure. Through time series analysis, it was found that the anomaly occurred within 30 seconds after each pump speed adjustment and lasted for about 10 seconds. This feature was classified into the blood circulation module, and it was found that the fluctuation of blood flow velocity was highly correlated with pressure anomaly. Further analysis found that the amplitude of pressure fluctuation was between -180mmHg and -220mmHg, which exceeded the lower limit of the standard range (-200-0mmHg). Spatial analysis of this anomaly was performed, and it was located that the anomaly mainly occurred in the pipeline section between the arterial end and the blood pump. Based on these analysis results, a series of control instructions were generated: reduce the pump speed by 20ml / min and observe the pressure change; if the pressure still did not return to normal, further check the position of the arterial end pipeline and adjust the puncture needle position if necessary. These control instructions are executed in priority order until the anomaly is resolved.

[0113] In a specific embodiment, the process of performing the multi-dimensional statistical analysis step may specifically include the following steps:

[0114] (1) Analyze the timestamp information in the abnormal operation data, calculate the frequency of abnormal events, and generate time dimension statistical data;

[0115] (2) Extracting the location information of system components from the abnormal operation data, counting the spatial distribution density of abnormal events, and obtaining spatial dimension statistical data;

[0116] (3) Classify the time dimension statistical data according to the anomaly type, calculate the temporal correlation of different types of anomalies, and generate anomaly type analysis data;

[0117] (4) Perform cluster analysis on spatial dimension statistical data to identify the spatial clustering characteristics of abnormal events and obtain fault area distribution data;

[0118] (5) Based on the abnormal type analysis data and the fault area distribution data, correlation analysis is performed to generate abnormal propagation path data;

[0119] (6) Generate equipment operation evaluation data based on the parameter change patterns and impact range in the abnormal propagation path data.

[0120] Specifically, the timestamp information of each abnormal event is extracted from the abnormal operation data, and the data is segmented according to a fixed time window, and the length of the time window is set to 5 minutes. The abnormal events in each time window are counted, and the number of occurrences, duration and time intervals of the abnormal events are recorded. The frequency analysis method is used to calculate the periodic characteristics of abnormal events, including short-term fluctuation frequency and long-term change trend, to form time dimension statistical data. The spatial dimension analysis is based on the location information of system components. The extracorporeal circulation system is divided into two major parts: the blood circulation path and the dialysate path. The blood circulation path is further subdivided into the arterial end, blood pump, dialyzer, venous end and other areas; the dialysate path is subdivided into water supply unit, concentrate preparation unit, waste liquid treatment unit and other areas. The abnormal events in each area are counted and counted, and the density of abnormal events in each area per unit time is calculated to form spatial dimension statistical data.

[0121] Abnormal type analysis adopts classification statistics method. Abnormal events in the time dimension statistical data are divided into pressure abnormality, blood flow abnormality, concentration abnormality and airtightness abnormality according to type. Time series correlation analysis is performed on each type of abnormal event to calculate the time interval and trigger relationship between different types of abnormalities. The time series dependency analysis method is used to identify the causal relationship between abnormal events and generate abnormal type analysis data. The cluster analysis of spatial dimension statistical data adopts density clustering algorithm. The distribution of abnormal events in space is represented as a density distribution map, and the abnormal density of each area is determined according to the frequency of occurrence of abnormal events. Then, the high-incidence and low-incidence areas of abnormal events are identified by density threshold division. The boundary division and feature extraction of high-incidence areas are performed to obtain the fault area distribution data.

[0122] The analysis of abnormal propagation paths is combined with spatiotemporal characteristics. The abnormal type analysis data is associated with the fault area distribution data to analyze the propagation order and impact range of abnormal events between different areas. By establishing a propagation model for abnormal events, the source location and diffusion path of the abnormality are identified, and the speed and range of abnormal diffusion are calculated to generate abnormal propagation path data. The generation of equipment operation evaluation data is based on the analysis results of the abnormal propagation path. The operating status of the equipment is evaluated by analyzing the change rules of abnormal parameters, including the speed, direction and range of parameter changes. At the same time, the impact degree and duration of the abnormality are considered to determine the priority and processing strategy of equipment maintenance. For example: In a hemodialysis treatment, the operation abnormality data showed that the arterial pressure fluctuated periodically. Time dimension analysis found that this fluctuation occurred every 15 minutes and lasted for about 2 minutes. Spatial positioning showed that the abnormality mainly occurred in the area between the arterial end and the blood pump. This abnormality was classified as a pressure abnormality type. Further analysis found that after each pressure abnormality occurred, the blood flow velocity would fluctuate accordingly within 30 seconds. Spatial clustering analysis determined that the high-incidence area of ​​the abnormality was the pipeline section from the arterial puncture needle to the blood pump inlet. Combined with the analysis of spatiotemporal characteristics, it was found that the abnormality first appeared at the arterial puncture needle, then spread along the pipeline to the blood pump, causing fluctuations in blood flow velocity. The evaluation showed that this abnormality was related to the decline in vascular access function, and it was necessary to adjust the puncture site or optimize the anticoagulation regimen.

[0123] The above describes the monitoring method of the extracorporeal circulation device in the hemodialysis intensive care unit in the embodiment of the present application. The following describes the monitoring system of the extracorporeal circulation device in the hemodialysis intensive care unit in the embodiment of the present application. Figure 2 In the embodiment of the present application, an embodiment of the extracorporeal circulation monitoring system of the hemodialysis intensive care unit includes:

[0124] The acquisition module 201 is used to collect vital signs and extract identity features of the patient through a multi-parameter monitoring device, obtain a patient vital sign feature data set through data fusion processing, and generate initial dialysis parameters based on the patient vital sign feature data set;

[0125] A detection module 202, configured to perform a vascular status analysis on the patient's physical sign feature data set to generate a vascular status parameter;

[0126] The adjustment module 203 is used to dynamically adjust the pressure value and blood flow value of the extracorporeal circulation system according to the vascular state parameters and the initial dialysis parameters, and obtain circulation control data through parameter correction;

[0127] A monitoring module 204 is used to monitor the water supply pressure, concentrate concentration and pipeline tightness of the dialysate circuit in real time based on the circulation control data, and obtain the circuit operation parameters through data analysis;

[0128] The analysis module 205 is used to perform multi-dimensional analysis on the system operation status according to the circulation control data and the fluid circuit operation parameters, and obtain the equipment early warning index by threshold comparison;

[0129] The control module 206 is used to perform statistical analysis on the operating parameters of the extracorporeal circulation device according to the device early warning indicators, and generate device operation evaluation data and automatic control instructions.

[0130] Through the collaboration of the above components, the patient's vital signs are collected and identity features are extracted through multi-parameter monitoring equipment. Combined with data fusion processing technology, the patient's biological information and physiological parameters are effectively integrated to form a comprehensive patient vital signs feature data set, avoiding the data island problem in traditional methods. The vascular state analysis based on the patient's vital signs feature data set innovatively applies non-invasive assessment methods, generates vascular state parameters through correlation analysis of physiological parameters such as blood pressure and heart rate, and improves the objectivity and accuracy of vascular state assessment. The mechanism of dynamic adjustment of the extracorporeal circulation system based on vascular state parameters and initial dialysis parameters realizes precise control of pressure values ​​and blood flow values. The circulation control data obtained through parameter correction provides a strong guarantee for the stable operation of the system and reduces the risk of complications during dialysis. The application of circulation control data further supports the real-time monitoring of the water supply pressure, concentrate concentration and pipeline tightness of the dialysis fluid circuit. The fluid circuit operation parameters obtained through data analysis improve the comprehensive monitoring of the equipment operation status. This solution innovatively combines circulation control data and fluid circuit operation parameters for multi-dimensional analysis. The equipment warning indicators generated by threshold comparison have high sensitivity and specificity, which significantly improves the early warning capability of abnormal situations. The equipment operation evaluation data and automatic control instructions generated based on the statistical analysis of the equipment warning indicators realize the self-optimization adjustment of the operation parameters of the extracorporeal circulator, reduce the workload of medical staff, and improve the safety and effectiveness of dialysis treatment. In particular, in the present invention, the application of artificial intelligence algorithms in data fusion, vascular status analysis and multi-dimensional anomaly detection enables the system to extract key features from massive monitoring data, identify potential abnormal patterns, and predict possible risks based on historical data and current status. The introduction of algorithm features significantly enhances the intelligence level of extracorporeal circulator monitoring, enables the system to have self-learning and self-adaptive capabilities, and can automatically adjust the monitoring strategy according to the individual differences of different patients, improve the accuracy and reliability of monitoring, and reduce operational errors by reducing human intervention.

[0131] The present application also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of the method for monitoring an extracorporeal circulator in a hemodialysis intensive care unit.

[0132] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0133] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.

[0134] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for monitoring an extracorporeal circulation device in a hemodialysis intensive care unit, characterized in that: The hemodialysis intensive care unit extracorporeal circulation device monitoring method comprises: The patient's vital signs are collected and identity features are extracted through a multi-parameter monitoring device, a patient's vital signs feature data set is obtained through data fusion processing, and initial dialysis parameters are generated based on the patient's vital signs feature data set; Performing vascular status analysis on the patient's physical sign feature data set to generate vascular status parameters; According to the vascular state parameters and the initial dialysis parameters, the pressure value and the blood flow value of the extracorporeal circulation system are dynamically adjusted, and circulation control data are obtained through parameter correction; Based on the circulation control data, the water supply pressure, concentrate concentration and pipeline tightness of the dialysis fluid circuit are monitored in real time, and the fluid circuit operation parameters are obtained through data analysis; According to the circulation control data and the fluid circuit operation parameters, a multi-dimensional analysis is performed on the system operation status, and the equipment early warning index is obtained by threshold comparison; Based on the equipment early warning indicators, statistical analysis is performed on the operating parameters of the extracorporeal circulation device to generate equipment operation evaluation data and automatic control instructions.

2. The method for monitoring an extracorporeal circulation device in a hemodialysis intensive care unit according to claim 1, characterized in that: The method includes collecting vital signs and extracting identity features of the patient through a multi-parameter monitoring device, obtaining a patient vital sign feature data set through data fusion processing, and generating initial dialysis parameters based on the patient vital sign feature data set, including: Collect facial features and fingerprint features of the patient through a biometric collection device to generate biometric data; The patient's heart rate, respiration, blood oxygen saturation, and blood pressure values ​​are sampled and acquired to generate physiological parameter data; Comparing and verifying the facial features and fingerprint features in the biometric data to obtain an identity authentication result; Compare the parameter values ​​in the physiological parameter data with the preset parameter ranges, wherein the heart rate parameter range is 50-130 times / min, the breathing parameter range is 10-30 times / min, the lower limit of the blood oxygen saturation parameter is 90%, the systolic pressure parameter range is 85-240 mmHg, and the diastolic pressure parameter range is 50-120 mmHg; Integrate the identity verification result and physiological parameter data to form a patient vital sign feature data set; The patient's weight data is collected by a weight sensor device, and the weight difference is calculated in combination with the physiological parameter data in the patient's physical sign characteristic data set, and the initial dialysis parameters are generated according to the weight difference.

3. The method for monitoring an extracorporeal circulation device in a hemodialysis intensive care unit according to claim 1, characterized in that: The step of performing vascular status analysis on the patient's physical sign feature data set to generate vascular status parameters includes: Extracting the blood pressure value and the heart rate value in the patient's physical sign feature data set, analyzing the ratio relationship between the blood pressure and the heart rate, calculating the blood pressure-heart rate multiplication factor, and generating vascular basic assessment data; Extracting pulse pressure difference and blood pressure fluctuation trend from the patient's physical sign feature data set, establishing a blood flow resistance evaluation index, and generating blood flow state prediction data; Performing correlation analysis on the body temperature value in the patient's physical sign feature data set and the blood oxygen saturation change in the physiological parameter data, constructing a relationship map between temperature and blood oxygen, and generating temperature correlation data; Performing segmented statistics on the resistance index in the blood flow state prediction data, calculating the fluctuation range to obtain a stability score, marking the fluctuation amplitude that exceeds a preset threshold, and generating blood flow stability data; Performing parameter correlation analysis on the temperature-related data and the blood flow stability data, calculating the vascular patency coefficient, and generating vascular access assessment data; Comprehensive weighted processing is performed on the vascular access assessment data and the vascular basic assessment data, and vascular status parameters are generated by data normalization.

4. The method for monitoring an extracorporeal circulation device in a hemodialysis intensive care unit according to claim 1, characterized in that: The method of dynamically adjusting the pressure value and blood flow value of the extracorporeal circulation system according to the vascular state parameters and the initial dialysis parameters, and obtaining circulation control data through parameter correction, includes: Merging the blood vessel state parameters with the initial dialysis parameters to generate initial adjustment parameters of the system; Extracting the arterial pressure threshold -200-0 mmHg, the venous pressure threshold 50-200 mmHg, and the transmembrane pressure threshold 0-400 mmHg from the initial adjustment parameters of the system to generate pressure control parameters; The real-time pressure value in the extracorporeal circulation system is collected by a pressure sensor, and compared with the pressure control parameter to obtain pressure deviation data; Based on the pressure deviation data, the blood flow velocity of the extracorporeal circulation system is divided into intervals, and adjusted according to a velocity gradient of 50 ml / min to generate blood flow adjustment data; Cross-validating the pressure deviation data with the blood flow regulation data, marking parameter points that exceed the safety range, and generating parameter correction data; The cycle control data is generated according to the corresponding relationship between the parameter correction data and the initial adjustment parameters of the system.

5. The method for monitoring an extracorporeal circulation device in a hemodialysis intensive care unit according to claim 1, characterized in that: Based on the circulation control data, the water supply pressure, concentrate concentration and pipeline tightness of the dialysis fluid circuit are monitored in real time, and the fluid circuit operation parameters are obtained through data analysis, including: Extracting pressure threshold and blood flow standard value from the circulation control data to generate fluid circuit monitoring benchmark data; The water supply pipeline pressure is sampled by a pressure sensor, and pressure fluctuation analysis is performed according to the liquid pipeline monitoring benchmark data to obtain water supply pressure data; Conduct conductivity test on dialysate concentrate, compare the conductivity value with the standard ratio to generate concentrate concentration data; The air pressure of each node of the pipeline is detected by the pressure sensor array, and the detection values ​​are arranged in time series to generate pipeline tightness data; Correlate and compare the water supply pressure data, the concentrate concentration data, and the pipeline tightness data, mark abnormal change points, and obtain abnormal liquid circuit data; The fluid circuit operation parameters are generated according to the comparison result between the fluid circuit abnormal data and the fluid circuit monitoring reference data.

6. The method for monitoring an extracorporeal circulation device in a hemodialysis intensive care unit according to claim 1, characterized in that: The system operation status is analyzed in multiple dimensions according to the circulation control data and the fluid circuit operation parameters, and the equipment early warning indicators are obtained by threshold comparison, including: Integrate the pressure value and blood flow value in the circulation control data with the water supply pressure, concentrate concentration, and pipeline tightness data in the fluid circuit operation parameters to obtain system operation data; Extracting pressure fluctuation values, blood flow stability values, concentration change values, and airtightness detection values ​​from the system operation data to generate parameter monitoring data; The parameter monitoring data is segmented according to the time series, and the data fluctuation range in each time period is compared with the standard threshold value to generate fluctuation analysis data; Extract features of abnormal points in the fluctuation analysis data through data analysis, calculate the change trends of various parameters, and obtain trend prediction data; Comparing the rate of change in the trend prediction data with a safety threshold, marking parameter points exceeding the threshold, and generating risk parameter data; Based on the time series distribution and change characteristics of the risk parameter data, equipment early warning indicators are generated.

7. The method for monitoring an extracorporeal circulation device in a hemodialysis intensive care unit according to claim 1, characterized in that: The method of performing statistical analysis on the operating parameters of the extracorporeal circulation device according to the equipment early warning indicators to generate equipment operation evaluation data and automatic control instructions includes: Extracting abnormal parameters from the equipment warning indicators, performing time series analysis on the changing rules of the parameters, and obtaining parameter change characteristic data; Classifying the parameter change characteristic data according to system function modules, calculating parameter correlation, and generating system performance evaluation data; Based on the system performance evaluation data, the fluctuation range of each operating parameter is counted to obtain parameter distribution data; Comparing the discrete points in the parameter distribution data with the standard operating parameters, marking the operating parameters whose deviation values ​​exceed the threshold, and generating abnormal operating data; Perform multi-dimensional statistical analysis based on the temporal and spatial distribution characteristics of the abnormal operation data to obtain equipment operation evaluation data; Automatic control instructions for the extracorporeal circulation device are generated by performing parameter optimization analysis on the device operation evaluation data.

8. The method for monitoring an extracorporeal circulation device in a hemodialysis intensive care unit according to claim 7, characterized in that: The multi-dimensional statistical analysis is performed according to the time distribution and spatial distribution characteristics of the abnormal operation data to obtain the equipment operation evaluation data, including: Analyze the timestamp information in the abnormal operation data, calculate the frequency of abnormal events, and generate time dimension statistical data; Extracting system component location information from the abnormal operation data, counting the spatial distribution density of abnormal events, and obtaining spatial dimension statistical data; Classify the time dimension statistical data according to the anomaly type, calculate the temporal correlation of different types of anomalies, and generate anomaly type analysis data; Performing cluster analysis on the spatial dimension statistical data to identify the spatial clustering characteristics of abnormal events and obtain fault area distribution data; Based on the abnormality type analysis data and the fault area distribution data, correlation analysis is performed to generate abnormality propagation path data; Equipment operation evaluation data is generated based on the parameter change rules and impact range in the abnormal propagation path data.

9. A monitoring system for an extracorporeal circulator in a hemodialysis intensive care unit, used to implement the monitoring method for an extracorporeal circulator in a hemodialysis intensive care unit as claimed in any one of claims 1 to 8, characterized in that: The extracorporeal circulation monitoring system of the hemodialysis intensive care unit comprises: The acquisition module is used to collect the patient's vital signs and extract identity features through a multi-parameter monitoring device, obtain a patient's vital signs feature data set through data fusion processing, and generate initial dialysis parameters based on the patient's vital signs feature data set; A detection module, used for performing vascular status analysis on the patient's physical sign feature data set to generate vascular status parameters; A regulating module, used for dynamically regulating the pressure value and blood flow value of the extracorporeal circulation system according to the vascular state parameters and the initial dialysis parameters, and obtaining circulation control data through parameter correction; A monitoring module, for real-time monitoring of the water supply pressure, concentrate concentration and pipeline tightness of the dialysate circuit based on the circulation control data, and obtaining the circuit operation parameters through data analysis; An analysis module, used to perform multi-dimensional analysis on the system operation status according to the circulation control data and the fluid circuit operation parameters, and obtain equipment early warning indicators through threshold comparison; The control module is used to perform statistical analysis on the operating parameters of the extracorporeal circulation device according to the equipment early warning indicators, and generate equipment operation evaluation data and automatic control instructions.

10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the method for monitoring an extracorporeal circulator in a hemodialysis intensive care unit according to any one of claims 1 to 8 is implemented.

Images