Hemodialysis equipment and method for nephrology department
By constructing a digital twin model and performing residual sequence analysis, the problem of traditional hemodialysis equipment's inability to detect dialyzer blockage in its early stages has been solved, achieving accurate fault detection and early warning.
Patent Information
- Application Number
- CN202511576479.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-31
- Publication Date
- 2026-03-03
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Traditional methods for detecting faults in hemodialysis equipment are insufficient to detect latent faults such as dialyzer blockage in the early stages, and they are also difficult to accurately pinpoint the root cause of the fault when multiple parameters are abnormal, leading to safety risks and impacting the adequacy of dialysis.
By constructing a digital twin model based on fluid dynamics principles, the system acquires blood pump speed and pressure sensor data, calculates theoretical values of arterial and venous pressure, extracts mean and trend slope features of the residual sequence, and determines dialyzer blockage faults.
It improves the specificity and anti-interference ability of fault detection, significantly enhances the accuracy of alarm information and its value in clinical decision support, and enables early fault diagnosis and status trend prediction.
Smart Images
Figure CN121594958A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of fault detection technology, and in particular to a hemodialysis device and method for nephrology. Background Technology
[0002] Fault inspection refers to the process of monitoring, testing and analyzing the status of equipment or systems through a series of means to determine whether there are functional abnormalities, performance degradation or impending failure. Its purpose is to discover and locate faults, covering everything from simple sensor threshold alarms to complex predictive diagnostics.
[0003] The auxiliary fault diagnosis of hemodialysis equipment in nephrology refers to an intelligent, model-driven fault diagnosis method applied to hemodialysis equipment in nephrology. Traditional hemodialysis equipment fault alarms are mostly based on fixed pressure thresholds. For slowly occurring "hidden faults" such as dialyzer blockage, alarms are only triggered when the pressure change accumulates to exceed the threshold. By this time, the fault has often progressed to the middle or late stage, which may have affected the adequacy of dialysis or posed a safety risk. Moreover, when multiple pressure parameters are abnormal at the same time, traditional methods are difficult to determine the root cause of the fault. For example, simultaneous increases in arterial and venous pressure may be due to dialyzer blockage, venous chamber blockage, or other reasons. The equipment cannot provide a clear fault indication. Therefore, how to detect faults in hemodialysis equipment based on the synergistic relationship between arterial and venous pressure has become a challenge for the industry. Summary of the Invention
[0004] Based on this, this application provides a hemodialysis device and method for nephrology that detects faults in hemodialysis equipment based on the synergistic relationship between arterial and venous pressure changes.
[0005] In a first aspect, this application provides a method for auxiliary examination of malfunctions in hemodialysis equipment used in nephrology, comprising the following steps: During the operation of the hemodialysis equipment, the actual value of the blood pump speed and the measured values of arterial pressure and venous pressure sensors are acquired simultaneously. The actual value of the blood pump speed is input into the pre-built digital twin model to obtain the corresponding theoretical values of the arterial pressure and the venous pressure. The residual sequence of arterial pressure between the sensor measured value and the corresponding theoretical value, and the residual sequence of venous pressure between the sensor measured value and the corresponding theoretical value are determined respectively. Temporal features are extracted from the arterial pressure residual sequence and the venous pressure residual sequence to obtain the corresponding mean features and trend slope features. Then, the faults of the hemodialysis equipment are judged based on the corresponding mean features and trend slope features. When the mean features corresponding to the arterial pressure residual sequence and the venous pressure residual sequence continuously exceed a preset positive threshold, and the trend slope features corresponding to the arterial pressure residual sequence and the venous pressure residual sequence both show a continuous positive increase, the hemodialysis equipment is determined to have a latent fault caused by dialyzer blockage, and the fault confidence of the latent fault is output to assist in the examination.
[0006] In some embodiments, the method specifically includes: pre-constructing a digital twin model of the hydraulic pressure in the extracorporeal blood circulation pathway of the hemodialysis device based on the principles of fluid dynamics, wherein the digital twin model is used to simulate the theoretical values of arterial pressure and venous pressure based on the blood pump speed and pipeline impedance characteristics of the hemodialysis device.
[0007] In some embodiments, constructing a digital twin model of the hydraulic pressure in the extracorporeal blood circulation pathway of a hemodialysis device based on fluid dynamics principles specifically includes: Obtain the flow resistance parameters of each section of the extracorporeal circulation tubing obtained through in vitro experiments; Establish the calculation relationship between blood pump speed and theoretical values of arterial and venous pressure based on fluid mechanics principles; The computational relationship is validated and corrected using clinical data to generate the final digital twin model.
[0008] In some embodiments, inputting the actual value of the blood pump rotation speed into a pre-built digital twin model to obtain the corresponding theoretical values of the arterial pressure and the venous pressure specifically includes: Use the actual value of the blood pump speed as the model input; Based on the rotational speed-flow rate relationship and pipeline flow resistance parameters in the pre-built digital twin model, the corresponding theoretical values of the arterial pressure and the venous pressure are calculated. Output the calculated theoretical values corresponding to the arterial pressure and the venous pressure.
[0009] In some embodiments, determining the residual sequence of arterial pressure between the sensor-measured value and the corresponding theoretical value specifically includes: Obtain the sensor measured values and corresponding theoretical values of the arterial pressure at each sampling time within a specified time period; The difference between the sensor-measured value and the corresponding theoretical value of arterial pressure at each sampling time is calculated to obtain the arterial pressure residual at each sampling time; All arterial pressure residuals are sorted according to the chronological order of sampling time to obtain the arterial pressure residual sequence.
[0010] In some embodiments, determining the venous pressure residual sequence between the sensor-measured value and the corresponding theoretical value of the venous pressure specifically includes: Obtain the sensor measured values and corresponding theoretical values of the venous pressure at each sampling time within a specified time period; The difference between the sensor-measured value and the corresponding theoretical value of venous pressure at each sampling time is calculated to obtain the venous pressure residual at each sampling time; All venous pressure residuals are sorted according to the chronological order of sampling time to obtain the venous pressure residual sequence.
[0011] In some embodiments, extracting time-domain features from the arterial pressure residual sequence and the venous pressure residual sequence to obtain corresponding mean features and trend slope features specifically includes: Mean features were calculated for the arterial pressure residual sequence and the venous pressure residual sequence, respectively. Trend slope features were calculated for the arterial pressure residual sequence and the venous pressure residual sequence, respectively. Output the mean characteristics and trend slope characteristics corresponding to the arterial pressure residual sequence and the venous pressure residual sequence.
[0012] Secondly, this application provides a hemodialysis device for nephrology, which includes a fault auxiliary inspection unit, the fault auxiliary inspection unit comprising: The acquisition module is used to simultaneously acquire the actual value of the blood pump speed and the measured values of arterial pressure and venous pressure from the sensors during the operation of the hemodialysis equipment. The processing module is used to input the actual value of the blood pump speed into a pre-built digital twin model to obtain the corresponding theoretical values of the arterial pressure and the venous pressure, and to determine the residual sequence of the arterial pressure between the sensor measured value and the corresponding theoretical value, and the residual sequence of the venous pressure between the sensor measured value and the corresponding theoretical value. The processing module is also used to extract time-domain features from the arterial pressure residual sequence and the venous pressure residual sequence to obtain the corresponding mean features and trend slope features, and then to make fault judgments on the hemodialysis equipment based on the corresponding mean features and trend slope features. The execution module is configured to determine that the hemodialysis device has a latent fault caused by dialyzer blockage when the mean features corresponding to the arterial pressure residual sequence and the venous pressure residual sequence continuously exceed a preset positive threshold, and the trend slope features corresponding to the arterial pressure residual sequence and the venous pressure residual sequence both show a continuous positive increase, and output the fault confidence of the latent fault to assist in the inspection.
[0013] Thirdly, this application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described auxiliary examination method for malfunctions of hemodialysis equipment for nephrology.
[0014] Fourthly, this application provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described auxiliary examination method for malfunctions of hemodialysis equipment for nephrology.
[0015] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects: In the hemodialysis equipment and method for nephrology provided in this application, firstly, the actual value of the blood pump rotation speed is input into a pre-built digital twin model to obtain the corresponding theoretical values of arterial and venous pressures. This step can generate a theoretical pressure benchmark based on physical principles, thereby effectively distinguishing between normal pressure fluctuations caused by individual patient differences and changes in treatment parameters and actual equipment failures. This provides a pure comparison benchmark unaffected by operating conditions for subsequent residual analysis, fundamentally improving the specificity and anti-interference capability of fault detection. Secondly, the residual order of the arterial pressure sensor measured value and the corresponding theoretical value is determined respectively. The system obtains the residual sequences of venous pressure, including the measured values from the sensor and the corresponding theoretical values. Then, it extracts time-domain features from both the arterial and venous pressure residual sequences to obtain corresponding mean and trend slope features. The mean feature characterizes the average amplitude level of the residual sequence within the statistical period, while the trend slope feature quantifies the direction and rate of change of the residual sequence over time. This can be used to dynamically capture and evaluate the evolution trend of abnormal system states. A positive growth trend indicates that the fault is worsening, while a negative decreasing trend indicates that the system may be recovering. Typically, this step transforms subtle fault information, hidden in the original pressure signal and difficult to detect directly, into explicit and quantifiable residual amplitude and trend characteristics. This allows for the capture of key dynamic information characterizing fault occurrence and evolution, providing a core basis for early fault diagnosis and state trend prediction, and improving the equipment's sensitivity to latent faults. Finally, when the mean characteristics corresponding to the arterial pressure residual sequence and the venous pressure residual sequence continuously exceed a preset positive threshold, and the trend slope characteristics corresponding to the arterial pressure residual sequence and the venous pressure residual sequence both show a continuous positive increase... The method determines that the hemodialysis equipment has a latent fault caused by dialyzer blockage and outputs the fault confidence level of the latent fault to assist in the inspection. This step can accurately associate and automatically identify specific characteristic patterns (double overthreshold + double positive increase) with the specific fault mechanism of "dialyzer blockage", and supplement it with quantitative confidence assessment. Thus, the general "high pressure alarm" can be upgraded to a specific and operable fault diagnosis conclusion, which significantly improves the accuracy of alarm information and clinical auxiliary decision-making value. In summary, the solution of this application can detect faults in hemodialysis equipment based on the synergistic change relationship between arterial pressure and venous pressure. Attached Figure Description
[0016] Figure 1 This is an exemplary flowchart of a method for assisting in the fault diagnosis of a hemodialysis device for nephrology, according to some embodiments of this application; Figure 2 This is a schematic diagram illustrating an application scenario of a fault-assisted inspection data processing system according to some embodiments of this application; Figure 3 This is a schematic flowchart illustrating the determination of arterial pressure residual sequences according to some embodiments of this application; Figure 4 This is a schematic diagram of the structure of a fault auxiliary inspection unit according to some embodiments of this application; Figure 5 This is a schematic diagram of the structure of a computer device for implementing a fault auxiliary examination method for a hemodialysis device used in nephrology, according to some embodiments of this application. Detailed Implementation
[0017] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.
[0018] refer to Figure 1 The figure is an exemplary flowchart of an auxiliary fault examination method for a nephrology hemodialysis device according to some embodiments of this application. The auxiliary fault examination method for a nephrology hemodialysis device mainly includes the following steps: In some embodiments, a digital twin model of the hydraulic pressure in the extracorporeal blood circulation pathway of the hemodialysis device is pre-constructed based on the principles of fluid mechanics. The digital twin model is used to simulate the theoretical values of arterial pressure and venous pressure based on the blood pump speed and pipeline impedance characteristics of the hemodialysis device.
[0019] In a preferred embodiment, constructing a digital twin model of the hydraulic pressure in the extracorporeal blood circulation pathway of a hemodialysis device based on fluid dynamics principles can be achieved through the following steps: Obtain the flow resistance parameters of each section of the extracorporeal circulation tubing obtained through in vitro experiments; Establish the calculation relationship between blood pump speed and theoretical values of arterial and venous pressure based on fluid mechanics principles; The computational relationship is validated and corrected using clinical data to generate the final digital twin model.
[0020] In practice, obtaining the flow resistance parameters of each segment of the extracorporeal circulation tubing calibrated through in vitro experiments can be achieved in the following way: First, a standard extracorporeal circulation loop can be constructed on the experimental bench using arterial tubing, dialyzer, and venous tubing of the same model as the clinical hemodialysis equipment. The test liquid preferably uses a 40% (w / w) glycerol aqueous solution with a viscosity similar to human blood dynamics as the blood simulation liquid to accurately simulate the flow resistance characteristics under clinical blood conditions. Then, the liquid is driven by a precision flow pump at at least five constant flow rates covering the normal operating range of the equipment, such as 0-500 mL / min, and the flow resistance parameters of the arterial tubing segment, dialyzer segment, and venous tubing at each flow rate point are recorded simultaneously. The stable differential pressure value of the pipeline section is measured by a high-precision pressure sensor. The pressure unit is uniformly in millimeters of mercury (mmHg), and the flow rate unit is uniformly in milliliters per minute (mL / min). Subsequently, based on Poiseuille's law in fluid mechanics, the differential pressure data measured for each pipeline at different flow rates is divided by the corresponding volumetric flow rate value to calculate the reference flow resistance of the corresponding pipeline under each effective operating condition. The flow resistance unit is mmHg·min / mL. Finally, the reference flow resistance under all effective operating conditions is arithmetically averaged to obtain the three key flow resistance parameters used in the digital twin model: the reference flow resistance of the arterial segment, the reference flow resistance of the dialyzer segment, and the reference flow resistance of the venous segment. Other methods can be used in other embodiments, and this application does not limit them.
[0021] In practical implementation, the calculation relationship between the blood pump rotation speed and the theoretical values of arterial and venous pressure based on fluid mechanics principles can be achieved as follows: First, the three flow resistance parameters obtained above are summed to obtain the total flow resistance of the extracorporeal circulation pathway, where the unit of total flow resistance is mmHg·min / ml; then, based on the rotation speed-flow characteristic curve calibrated at the factory of the blood pump, a linear interpolation lookup table is established to map the input rotation speed to the theoretical output flow rate, where the unit of rotation speed is revolutions per minute and the unit of theoretical output flow rate is ml / min. During the digital twin model calculation, for any input actual value of the blood pump rotation speed, the corresponding theoretical volumetric flow rate can be obtained by querying this linear interpolation lookup table; Then, based on the principle that the total pressure drop of the series hydraulic circuit is equal to the product of the total flow resistance and the flow rate, the theoretical flow rate is multiplied by the total flow resistance to calculate the theoretical total pressure drop from the arterial end to the venous end. Finally, based on the proportion of the arterial segment flow resistance and the venous side flow resistance in the total flow resistance, the theoretical total pressure drop is proportionally distributed to the arterial end and the venous end. The theoretical value of the arterial pressure is the pressure drop generated by the arterial segment flow resistance before the blood pump inlet, which is usually a negative value. The theoretical value of the venous pressure is the sum of the pressure drops generated by the dialyzer segment and the venous segment flow resistance after the blood pump outlet, which is usually a positive value. This completes the mapping calculation from the blood pump speed to the theoretical values of the arterial and venous pressures. Other methods can be used in other embodiments, and this application does not limit them.
[0022] In specific implementation, the final digital twin model can be generated by verifying and correcting the calculation relationship through clinical data in the following manner: During the clinical trial phase of the hemodialysis device, multiple sets of measured values of arterial and venous pressure sensors at different blood pump speeds are anonymously collected during patient treatment; the collected blood pump speed data are input into the initial digital twin model to obtain the corresponding theoretical pressure value, which is then compared with the clinically measured pressure value; when the average error between the theoretical value and the clinically measured value of the initial digital twin model exceeds a preset allowable range, such as 10%, a linear regression algorithm is used to fine-tune and optimize the reference flow resistance parameters of each tube segment in the initial digital twin model so that the curve of the theoretical pressure value output by the initial digital twin model achieves the best fit with the distribution trend of the clinically measured pressure data; finally, after the clinical data verification and parameter correction process, the optimized and finalized flow resistance parameters and calculation relationships are solidified into the final digital twin model. Other methods can also be used in other embodiments, and this application does not limit them.
[0023] In step 101, during the operation of the hemodialysis equipment, the actual value of the blood pump speed and the measured values of the arterial pressure and venous pressure sensors are acquired simultaneously.
[0024] In specific implementation, during the operation of the hemodialysis equipment, the actual value of the blood pump speed and the measured values of the arterial and venous pressure sensors can be obtained simultaneously in the following way: During the operation of the hemodialysis equipment, the encoder pulse signal from the blood pump servo motor can be read concurrently at a fixed sampling frequency of 10 Hz to calculate the actual value of the blood pump speed, and the output voltage signal of the silicon piezoresistive pressure sensor installed on the arterial and venous chambers can be acquired simultaneously through its integrated multi-channel high-precision analog-to-digital converter; then, a second-order Butterworth low-pass digital filter with a cutoff frequency of 2 Hz is applied to the acquired output voltage signal to suppress high-frequency noise introduced by tubing vibration and blood pump pulsation, thereby obtaining the measured values of the arterial and venous pressure sensors after noise suppression. Other methods can also be used in other embodiments, which are not limited in this application.
[0025] In some embodiments, reference Figure 2 As shown in the figure, this figure is a schematic diagram of the application scenario of the fault auxiliary inspection data processing system shown in some embodiments of this application. The figure includes three main components: acquisition device, server and data storage device. The acquisition device is responsible for collecting the actual value of the blood pump speed of the hemodialysis equipment and the actual measured values of the arterial pressure and venous pressure sensors. It sends the collected actual value of the blood pump speed and the actual measured values of the arterial pressure and venous pressure sensors to the server through the communication network. The fault auxiliary inspection data processing system runs on the server. The server stores the processing results in the data storage device and visualizes them.
[0026] In step 102, the actual value of the blood pump speed is input into the pre-built digital twin model to obtain the corresponding theoretical values of the arterial pressure and the venous pressure. The residual sequence of the arterial pressure between the sensor measured value and the corresponding theoretical value, and the residual sequence of the venous pressure between the sensor measured value and the corresponding theoretical value are determined respectively.
[0027] In some embodiments, inputting the actual value of the blood pump rotation speed into a pre-built digital twin model to obtain the corresponding theoretical values of the arterial pressure and the venous pressure can be achieved by the following steps: Use the actual value of the blood pump speed as the model input; Based on the rotational speed-flow rate relationship and pipeline flow resistance parameters in the pre-built digital twin model, the corresponding theoretical values of the arterial pressure and the venous pressure are calculated. Output the calculated theoretical values corresponding to the arterial pressure and the venous pressure.
[0028] In specific implementation, based on the rotational speed-flow rate relationship and pipeline flow resistance parameters in the pre-built digital twin model, the corresponding theoretical values of the arterial pressure and the venous pressure can be calculated in the following way: The internal calculation logic of the digital twin model first maps the received actual value of the blood pump rotational speed to the corresponding theoretical volumetric flow rate value according to the linear interpolation lookup table stored in it and pre-calibrated through experiments; then, the digital twin model calls the reference flow resistance parameters of the arterial segment, dialyzer segment and venous segment stored in it, and calculates the corresponding theoretical value of the arterial pressure and the theoretical value of the venous pressure according to the pressure drop calculation principle of the series hydraulic circuit. The corresponding theoretical value of the arterial pressure is used as the pressure drop before the blood pump inlet, and the theoretical value of the venous pressure is used as the total pressure drop after the blood pump outlet. Other methods can also be used in other embodiments, and this application does not limit them.
[0029] It should be noted that the above steps can generate a theoretical pressure benchmark based on physical principles, which can effectively distinguish between normal pressure fluctuations caused by individual patient differences and changes in treatment parameters and actual equipment failures. This provides a pure comparison benchmark that is not affected by operating conditions for subsequent residual analysis, fundamentally improving the specificity and anti-interference capability of fault detection.
[0030] In some embodiments, reference Figure 3 As shown in the figure, this is a flowchart illustrating the process of determining the arterial pressure residual sequence in some embodiments of this application. In this embodiment, determining the arterial pressure residual sequence between the sensor-measured value and the corresponding theoretical value of the arterial pressure can be achieved through the following steps: In step 1031, the sensor measured values and corresponding theoretical values of the arterial pressure at each sampling time within a specified time period are obtained; In step 1032, the difference between the sensor-measured value and the corresponding theoretical value of arterial pressure at each sampling time is calculated to obtain the arterial pressure residual at each sampling time; In step 1033, all arterial pressure residuals are sorted according to the time sequence of sampling time to obtain the arterial pressure residual sequence.
[0031] In specific implementation, the sensor measured values and corresponding theoretical values of the arterial pressure at each sampling moment within a specified time period can be obtained in the following way: data units of all sampling moments within a specified time period can be read from the hemodialysis device. The specified time period is, for example, the time period from the current moment to the previous 30 seconds. Each data unit of sampling moment includes a unified timestamp, a sensor measured value of the arterial pressure collected from the arterial pressure sensor at the corresponding timestamp and filtered, and a theoretical value of the arterial pressure calculated and output by the digital twin model at the same timestamp. Other methods can also be used in other embodiments, and this application does not limit them.
[0032] In specific implementation, the difference between the sensor-measured value and the corresponding theoretical value of arterial pressure at each sampling time is calculated to obtain the arterial pressure residual at each sampling time. This can be achieved in the following way: for each sampling time, the difference between the sensor-measured value and the corresponding theoretical value of arterial pressure at the sampling time is taken as the arterial pressure residual at the corresponding sampling time. Other methods can also be used in other embodiments, and this application does not limit them.
[0033] In some embodiments, determining the venous pressure residual sequence between the sensor-measured value and the corresponding theoretical value of the venous pressure can be achieved by the following steps: Obtain the sensor measured values and corresponding theoretical values of the venous pressure at each sampling time within a specified time period; The difference between the sensor-measured value and the corresponding theoretical value of venous pressure at each sampling time is calculated to obtain the venous pressure residual at each sampling time; All venous pressure residuals are sorted according to the chronological order of sampling time to obtain the venous pressure residual sequence.
[0034] In specific implementation, the sensor measured values and corresponding theoretical values of venous pressure at each sampling moment within a specified time period can be obtained in the following way: data units of all sampling moments within a specified time period can be read from the hemodialysis device. The specified time period is, for example, the time period from the current moment to the previous 30 seconds. Each data unit of sampling moment includes a unified timestamp, a sensor measured value of venous pressure collected from the venous pressure sensor at the corresponding timestamp and filtered, and a theoretical value of venous pressure calculated and output by the digital twin model at the same timestamp. Other methods can also be used in other embodiments, and this application does not limit them.
[0035] In specific implementation, the difference between the sensor measured value and the corresponding theoretical value of venous pressure at each sampling time is calculated to obtain the venous pressure residual at each sampling time. This can be achieved in the following way: for each sampling time, the difference between the sensor measured value and the corresponding theoretical value of venous pressure at the sampling time is taken as the venous pressure residual at the corresponding sampling time. Other methods can also be used in other embodiments, and this application does not limit them.
[0036] It should be noted that, in this application, the arterial pressure residual refers to the difference between the actual measured value of arterial pressure by the sensor at a specific sampling moment and the corresponding theoretical value calculated by its digital twin model; the venous pressure residual refers to the difference between the actual measured value of venous pressure by the sensor at a specific sampling moment and the corresponding theoretical value calculated by its digital twin model; the core function of the arterial pressure residual and the venous pressure residual is to quantify the instantaneous deviation between the actual state of the hydraulic system monitored in real time and the expected state of the ideal model.
[0037] In step 103, time-domain features are extracted from the arterial pressure residual sequence and the venous pressure residual sequence to obtain the corresponding mean features and trend slope features. Then, the fault judgment of the hemodialysis equipment is made based on the corresponding mean features and trend slope features.
[0038] In some embodiments, extracting time-domain features from the arterial pressure residual sequence and the venous pressure residual sequence to obtain the corresponding mean features and trend slope features can be achieved by the following steps: Mean features were calculated for the arterial pressure residual sequence and the venous pressure residual sequence, respectively. Trend slope features were calculated for the arterial pressure residual sequence and the venous pressure residual sequence, respectively. Output the mean characteristics and trend slope characteristics corresponding to the arterial pressure residual sequence and the venous pressure residual sequence.
[0039] In specific implementation, the mean feature calculation for the arterial pressure residual sequence and the venous pressure residual sequence can be achieved in the following way: For the arterial pressure residual sequence and the venous pressure residual sequence within a specified time window of 30 seconds, the arithmetic mean of all residual values within the time window can be calculated. Specifically, for the arterial pressure residual sequence, the arterial pressure residuals at each sampling time are summed, and then divided by the total number of sampling times of the arterial pressure residual sequence. The resulting value is the mean feature of the arterial pressure residual sequence within the specified time period, reflecting the average amplitude level of the arterial pressure residuals within the statistical period. For the venous pressure residual sequence, the venous pressure residuals at each sampling time are summed, and then divided by the total number of sampling times of the venous pressure residual sequence. The resulting value is the mean feature of the venous pressure residual sequence within the specified time period, reflecting the average amplitude level of the venous pressure residuals within the statistical period. Other methods can also be used in other embodiments, and this application does not limit them.
[0040] It should be noted that the mean characteristic in this application characterizes the average amplitude level of the residual sequence within the statistical period. It refers to a statistic obtained by averaging all residual values in the arterial pressure residual sequence or venous pressure residual sequence within a specified time window. It can be used to filter out instantaneous fluctuation interference and thus reflect the persistence and stability of pressure deviation from its theoretical benchmark. A higher mean characteristic indicates that there is a persistent static deviation in the system pressure.
[0041] In specific implementation, the trend slope feature calculation for the arterial pressure residual sequence and the venous pressure residual sequence can be achieved in the following way: The least squares method can be used to perform linear trend fitting on the arterial pressure residual sequence and the venous pressure residual sequence within a specified time period of 30 seconds. More specifically, for the arterial pressure residual sequence, with the sampling time point as the independent variable and the arterial pressure residual as the dependent variable, an optimal fitting line is obtained. The slope of this fitting line is defined as the trend slope feature of the arterial pressure residual sequence, which quantifies the direction and rate of change of the arterial pressure residual within the statistical period. For the venous pressure residual sequence, with the sampling time point as the independent variable and the venous pressure residual as the dependent variable, an optimal fitting line is obtained. The slope of this fitting line is defined as the trend slope feature of the venous pressure residual sequence, which quantifies the direction and rate of change of the venous pressure residual within the statistical period. Other methods can also be used in other embodiments, and this application does not limit them.
[0042] It should be noted that the trend slope feature in this application quantifies the direction and rate of change of the residual sequence over time. It refers to the slope of the fitted line obtained by performing trend analysis on the arterial pressure residual sequence or venous pressure residual sequence within a specified time window using a linear fitting method. This slope can be used to dynamically capture and evaluate the evolution trend of abnormal system states. A positive growth trend indicates that the fault is developing and deteriorating, while a negative decreasing trend indicates that the system may be recovering to normal.
[0043] In addition, it should be noted that the above steps can transform the subtle fault information hidden in the original pressure signal that is difficult to detect directly into explicit and quantifiable residual amplitude and trend characteristics. This allows for the capture of key dynamic information characterizing the occurrence and evolution of faults, providing a core basis for early fault diagnosis and state trend prediction, and improving the equipment's sensitivity to latent faults.
[0044] In step 104, when the mean features corresponding to the arterial pressure residual sequence and the venous pressure residual sequence continuously exceed a preset positive threshold, and the trend slope features corresponding to the arterial pressure residual sequence and the venous pressure residual sequence both show a continuous positive increase, the hemodialysis device is determined to have a latent fault caused by dialyzer blockage, and the fault confidence of the latent fault is output to assist in the inspection.
[0045] It should be noted that the "continuous" judgment is implemented through a sliding time window mechanism. More specifically, a judgment window with a length of 30 seconds is set, and it slides once every 5 seconds. When the following two conditions are met simultaneously within 3 consecutive judgment windows, the dialyzer blockage fault judgment is triggered.
[0046] Additionally, it should be noted that the positive thresholds corresponding to the arterial pressure residual sequence and the venous pressure residual sequence in this application can be determined using the following method: The positive thresholds corresponding to the arterial pressure residual sequence and the venous pressure residual sequence can be determined using a method based on the statistical distribution of normal clinical data. More specifically, firstly, during the clinical trial phase of the hemodialysis device, a large amount of mean characteristic data of the arterial pressure residual sequence and the venous pressure residual sequence under normal operating conditions (i.e., no faults and stable treatment) are anonymously collected. Subsequently, the distribution statistical analysis is performed on all the collected mean characteristic data under normal operating conditions, and their statistical quantiles are calculated. Finally, the positive threshold of the arterial pressure residual sequence is set to the 95th quantile value corresponding to the normal data distribution, and the positive threshold of the venous pressure residual sequence is set to the 95th quantile value corresponding to the normal data distribution. This method ensures that within the normal fluctuation range, only 5% of the extreme values will exceed the corresponding positive threshold, thereby providing a highly specific statistical benchmark for fault detection. In other embodiments, other quantiles such as 90% or 99% can also be used, and this application does not limit this.
[0047] In specific implementation, when the hemodialysis equipment is determined to have a latent fault caused by dialyzer blockage, the fault confidence of the latent fault can be determined in the following way: the fault confidence can be calculated based on the mean and trend slope characteristics corresponding to the arterial pressure residual sequence and the venous pressure residual sequence within the current specified time period. For example, the calculation method for the fault confidence is as follows: first, the mean and trend slope characteristics corresponding to the arterial pressure residual sequence and the venous pressure residual sequence are normalized respectively, mapping the corresponding mean and trend slope characteristics to the [0, 1] interval. Then, the normalized mean and trend slope characteristics are substituted into a predefined weighted summation formula (e.g., confidence = w1*Norm(μ_a) + w2*Norm(μ_a)). The formula is: μ_v) + w3*Norm(k_a) + w4*Norm(k_v)), where w1 to w4 are preset weight coefficients, Norm(μ_a) represents the result after normalizing the mean feature of the arterial pressure residual sequence, Norm(μ_v) represents the result after normalizing the mean feature of the venous pressure residual sequence, Norm(k_a) represents the result after normalizing the trend slope feature of the arterial pressure residual sequence, and Norm(k_v) represents the result after normalizing the trend slope feature of the venous pressure residual sequence. Finally, the calculated result is mapped to a value between 0% and 100% as the fault confidence. Other methods can be used in other embodiments, and this application does not limit them.
[0048] It should be noted that the normalization method in this application can adopt linear normalization methods such as maximum and minimum value scaling; the weight coefficients w1 to w4 can be initially assigned values based on professional knowledge in the field of hemodialysis and fluid dynamics principles. Since dialyzer blockage faults are mainly manifested in the hydraulic system as a significant increase in arterial and venous pressure values, the mean characteristic reflecting the pressure deviation from the average level is usually given a higher weight. For example, w1 and w2 can be initially set to 0.4 each, while w3 and w4 can be set to 0.1 each, so that the static deviation of pressure is given more weight in fault determination. This set of initial weights can be used as baseline parameters. Other empirical allocation ratios can also be used in other embodiments, and this application does not limit them.
[0049] In some embodiments, the fault diagnosis of the hemodialysis equipment based on the corresponding mean characteristics and trend slope characteristics further includes the identification of other fault modes. Specifically, a fault feature mode library is constructed, which pre-stores the mapping relationship between different fault types and residual feature combinations, i.e.: When the mean features corresponding to the arterial pressure residual sequence and the venous pressure residual sequence do not exceed the preset positive threshold, but the trend slope features corresponding to the arterial pressure residual sequence and the venous pressure residual sequence show a statistically significant and continuous positive increase in multiple consecutive time windows, the hemodialysis device is determined to be in an early latent flow resistance increase state. When the mean features corresponding to the arterial pressure residual sequence and the venous pressure residual sequence continuously exceed the preset positive threshold, but the trend slope features corresponding to the arterial pressure residual sequence and the venous pressure residual sequence show a statistically significant continuous negative decrease in multiple consecutive time windows, the hemodialysis equipment is determined to be in fault recovery. When the mean features corresponding to the arterial pressure residual sequence and the venous pressure residual sequence do not exceed the preset positive threshold, but the trend slope features corresponding to the arterial pressure residual sequence and the venous pressure residual sequence show a statistically significant and continuous negative decrease in multiple consecutive time windows, the hemodialysis equipment is determined to be in normal condition. When the mean value of the arterial pressure residual sequence continuously exceeds the preset negative threshold and its trend slope shows a continuous negative increase, while the characteristic change of the venous pressure residual sequence is not significant, the failure mode of the hemodialysis equipment is determined to be arterial end tubing collapse or arterial chamber blockage. When the mean value of the venous pressure residual sequence continuously exceeds a preset positive threshold and its trend slope shows a continuous positive increase, while the characteristic change of the arterial pressure residual sequence is not significant, the fault mode of the hemodialysis equipment is determined to be venous pot or venous end tubing blockage.
[0050] It should be noted that the above steps can accurately associate and automatically identify specific characteristic patterns (double overthreshold + double positive increase) with the specific fault mechanism of "dialyzer blockage", and supplement them with quantitative confidence assessment. This can upgrade the general "high pressure alarm" into a specific and actionable fault diagnosis conclusion, which significantly improves the accuracy of alarm information and its clinical decision support value.
[0051] Furthermore, in another aspect of this application, in some embodiments, this application provides a hemodialysis device for nephrology, which includes a fault auxiliary inspection unit, as referenced. Figure 4 The figure is a schematic diagram of the structure of a fault auxiliary inspection unit according to some embodiments of this application. The fault auxiliary inspection unit includes: an acquisition module 401, a processing module 402, and an execution module 403, which are described below: The acquisition module 401 in this application is mainly used to simultaneously acquire the actual value of the blood pump speed and the measured values of the arterial pressure and venous pressure sensors during the operation of the hemodialysis equipment. Processing module 402, in this application, is mainly used to input the actual value of the blood pump speed into a pre-built digital twin model to obtain the corresponding theoretical values of the arterial pressure and the venous pressure, and to determine the residual sequence of the arterial pressure between the sensor measured value and the corresponding theoretical value, and the residual sequence of the venous pressure between the sensor measured value and the corresponding theoretical value. The processing module 402 described in this application is also used to extract time-domain features from the arterial pressure residual sequence and the venous pressure residual sequence to obtain the corresponding mean features and trend slope features, and then to make fault judgments on the hemodialysis equipment based on the corresponding mean features and trend slope features. The execution module 403 in this application is mainly used to determine that the hemodialysis equipment has a latent fault caused by dialyzer blockage when the mean features corresponding to the arterial pressure residual sequence and the venous pressure residual sequence continuously exceed a preset positive threshold, and the trend slope features corresponding to the arterial pressure residual sequence and the venous pressure residual sequence both show a continuous positive increase. The execution module 403 is mainly used to determine that the hemodialysis equipment has a latent fault caused by dialyzer blockage, and output the fault confidence of the latent fault to assist in the inspection.
[0052] Each module in the aforementioned fault auxiliary inspection unit can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the corresponding operations of each module.
[0053] In another embodiment, this application provides a computer device, which may be a server, and its internal structure diagram may be as follows. Figure 5 As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computational and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores auxiliary diagnostic data for nephrology hemodialysis equipment. The network interface communicates with external terminals via a network connection. When executed by the processor, the computer program implements a method for auxiliary diagnostic testing of nephrology hemodialysis equipment.
[0054] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.
[0055] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above embodiment of the auxiliary examination method for fault diagnosis of hemodialysis equipment for nephrology.
[0056] In one embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described embodiment of the method for auxiliary examination of faults in a hemodialysis device for nephrology.
[0057] In one embodiment, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the steps described in the embodiment of the method for auxiliary examination of malfunctions in a nephrology hemodialysis device.
[0058] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.
[0059] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0060] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.
Claims
1. A method for auxiliary examination of malfunctions in hemodialysis equipment used in nephrology, characterized in that, Includes the following steps: During the operation of the hemodialysis equipment, the actual value of the blood pump speed and the measured values of arterial pressure and venous pressure sensors are acquired simultaneously. The actual value of the blood pump speed is input into the pre-built digital twin model to obtain the corresponding theoretical values of the arterial pressure and the venous pressure. The residual sequence of arterial pressure between the sensor measured value and the corresponding theoretical value, and the residual sequence of venous pressure between the sensor measured value and the corresponding theoretical value are determined respectively. Temporal features are extracted from the arterial pressure residual sequence and the venous pressure residual sequence to obtain the corresponding mean features and trend slope features. Then, the faults of the hemodialysis equipment are judged based on the corresponding mean features and trend slope features. When the mean features corresponding to the arterial pressure residual sequence and the venous pressure residual sequence continuously exceed a preset positive threshold, and the trend slope features corresponding to the arterial pressure residual sequence and the venous pressure residual sequence both show a continuous positive increase, the hemodialysis equipment is determined to have a latent fault caused by dialyzer blockage, and the fault confidence of the latent fault is output to assist in the examination.
2. The method as described in claim 1, characterized in that, Specifically, it includes: A digital twin model of the hydraulic pressure in the extracorporeal blood circulation pathway of a hemodialysis device is pre-constructed based on the principles of fluid mechanics. The digital twin model is used to simulate the theoretical values of arterial and venous pressures based on the blood pump speed and pipeline impedance characteristics of the hemodialysis device.
3. The method as described in claim 2, characterized in that, The digital twin model of hydraulic pressure in the extracorporeal blood circulation pathway of hemodialysis equipment, based on the principles of fluid mechanics, specifically includes: Obtain the flow resistance parameters of each section of the extracorporeal circulation tubing obtained through in vitro experiments; Establish the calculation relationship between blood pump speed and theoretical values of arterial and venous pressure based on fluid mechanics principles; The computational relationship is validated and corrected using clinical data to generate the final digital twin model.
4. The method as described in claim 1, characterized in that, The actual value of the blood pump rotation speed is input into a pre-built digital twin model to obtain the corresponding theoretical values of the arterial pressure and the venous pressure, specifically including: Use the actual value of the blood pump speed as the model input; Based on the rotational speed-flow rate relationship and pipeline flow resistance parameters in the pre-built digital twin model, the corresponding theoretical values of the arterial pressure and the venous pressure are calculated. Output the calculated theoretical values corresponding to the arterial pressure and the venous pressure.
5. The method as described in claim 1, characterized in that, The process of determining the residual sequence of arterial pressure between the sensor's measured value and the corresponding theoretical value specifically includes: Obtain the sensor measured values and corresponding theoretical values of the arterial pressure at each sampling time within a specified time period; The difference between the sensor-measured value and the corresponding theoretical value of arterial pressure at each sampling time is calculated to obtain the arterial pressure residual at each sampling time; All arterial pressure residuals are sorted according to the chronological order of sampling time to obtain the arterial pressure residual sequence.
6. The method as described in claim 1, characterized in that, The venous pressure residual sequence, which determines the sensor measured value and the corresponding theoretical value of the venous pressure, specifically includes: Obtain the sensor measured values and corresponding theoretical values of the venous pressure at each sampling time within a specified time period; The difference between the sensor-measured value and the corresponding theoretical value of venous pressure at each sampling time is calculated to obtain the venous pressure residual at each sampling time; All venous pressure residuals are sorted according to the chronological order of sampling time to obtain the venous pressure residual sequence.
7. The method as described in claim 1, characterized in that, Temporal feature extraction is performed on the arterial pressure residual sequence and the venous pressure residual sequence to obtain the corresponding mean feature and trend slope feature, specifically including: Mean features were calculated for the arterial pressure residual sequence and the venous pressure residual sequence, respectively. Trend slope features were calculated for the arterial pressure residual sequence and the venous pressure residual sequence, respectively. Output the mean characteristics and trend slope characteristics corresponding to the arterial pressure residual sequence and the venous pressure residual sequence.
8. A hemodialysis device for nephrology, the device comprising a fault-tolerant auxiliary inspection unit, characterized in that, The fault auxiliary inspection unit includes: The acquisition module is used to simultaneously acquire the actual value of the blood pump speed and the measured values of arterial pressure and venous pressure from the sensors during the operation of the hemodialysis equipment. The processing module is used to input the actual value of the blood pump speed into a pre-built digital twin model to obtain the corresponding theoretical values of the arterial pressure and the venous pressure, and to determine the residual sequence of the arterial pressure between the sensor measured value and the corresponding theoretical value, and the residual sequence of the venous pressure between the sensor measured value and the corresponding theoretical value. The processing module is also used to extract time-domain features from the arterial pressure residual sequence and the venous pressure residual sequence to obtain the corresponding mean features and trend slope features, and then to make fault judgments on the hemodialysis equipment based on the corresponding mean features and trend slope features. The execution module is configured to determine that the hemodialysis device has a latent fault caused by dialyzer blockage when the mean features corresponding to the arterial pressure residual sequence and the venous pressure residual sequence continuously exceed a preset positive threshold, and the trend slope features corresponding to the arterial pressure residual sequence and the venous pressure residual sequence both show a continuous positive increase, and output the fault confidence of the latent fault to assist in the inspection.
9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the auxiliary fault inspection method for a hemodialysis device for nephrology as described in any one of claims 1 to 7.
10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for auxiliary fault examination of hemodialysis equipment for nephrology as described in any one of claims 1 to 7.