Intelligent parameter configuration method and system for hemodialysis device

Abstract

The invention relates to the technical field of hemodialysis dynamic regulation and control, in particular to an intelligent parameter configuration method and system for a hemodialysis device. The method comprises the following steps: firstly, acquiring an initial capacity response coefficient and a relaxed pulsation amplitude; further extracting a pipe wall shrinkage ratio, and obtaining a corrected capacity response coefficient in combination with the initial capacity response coefficient; further based on the instantaneous change of the capacity attenuation fundamental wave, combining with the corrected capacity response coefficient, and comparing with the real-time removal rate of the ultrafiltration pump to obtain the blood vessel recharging rate; further obtaining a hemodynamic load index; and finally, obtaining a current rigid demand rate based on the global task state variable. Parameters of the hemodialysis device are configured according to the current hemodynamics load index, the rigidity demand rate and the blood vessel recharging rate, the problem that due to vasoconstriction, estimation of the recharging rate by a traditional linear model is deviated is effectively solved, and conversion from lag feedback to prospective self-adaptive control is achieved.

Description

Technical Field

[0001] This invention relates to the field of dynamic regulation technology for hemodialysis, specifically to a method and system for intelligent parameter configuration of a hemodialysis device. Background Technology

[0002] During hemodialysis treatment, the core objective of ultrafiltration control is to remove excess water accumulated in the patient's body within a limited treatment time, while avoiding complications such as hypotension caused by excessively rapid removal. Current dialysis equipment typically relies on online blood volume monitoring technology to measure the trend of relative blood volume (RBV) changes and adjust the ultrafiltration rate accordingly.

[0003] However, existing control strategies typically assume that vascular compliance remains constant during treatment, directly adjusting the ultrafiltration rate based on the slope of the relative blood volume decrease. In reality, when effective circulating blood volume decreases due to ultrafiltration, the body undergoes compensatory vasoconstriction, leading to decreased vascular compliance (i.e., vascular hardening). At this point, even if the interstitial fluid refill rate is adequate, hardened vessels can cause the relative blood volume reading to show a more dramatic decrease than the actual water loss. Current technology ignores this physical change caused by vasoconstriction, easily misinterpreting this accelerated decrease in relative blood volume due to decreased vascular compliance as insufficient tissue refill or excessive dehydration. This leads to incorrectly limiting the ultrafiltration rate, resulting in reduced treatment efficiency and difficulty in accurately matching the patient's true physiological capacity. Summary of the Invention

[0004] To address the technical problem of inaccurate ultrafiltration control caused by neglecting compliance changes resulting from vasoconstriction in existing technologies, the present invention aims to provide a method and system for intelligent parameter configuration of a hemodialysis device. The specific technical solution adopted is as follows: A method for intelligent parameter configuration of a hemodialysis device, the method comprising: Within the preset initial calibration time window, the initial capacity response coefficient and relaxation-state pulsation amplitude are obtained; Based on the fluctuations of the vessel wall pulsation wave data within the current preset neighborhood window, and combined with the relaxation state pulsation amplitude, the vessel wall contraction ratio is obtained; the current vessel wall contraction ratio and the initial capacity response coefficient are fused to obtain the corrected capacity response coefficient; based on the instantaneous change of the capacity decay fundamental wave, and combined with the corrected capacity response coefficient, the real-time removal rate of the ultrafiltration pump is compared to obtain the vessel refill rate. By comparing the current vascular refill rate and the real-time removal rate, and combining the distribution of the volume decay fundamental wave, the hemodynamic load index is obtained; the current rigid demand rate is obtained based on the global task state variables; and the parameters of the hemodialysis device are configured according to the current hemodynamic load index, the rigid demand rate, and the vascular refill rate.

[0005] Furthermore, the method for obtaining the hemodynamic load index includes: By comparing the current vascular refill rate and the real-time removal rate, a dynamic supply-demand factor is obtained; based on the degree to which the fundamental value of capacity decay approaches a preset capacity threshold, a static inventory factor is obtained; and by fusing the current dynamic supply-demand factor and the static inventory factor, a hemodynamic load index is obtained.

[0006] Furthermore, the method for obtaining the rigid demand rate includes: Based on the preset total ultrafiltration target and the current and previous real-time removal rates, the remaining task volume is obtained; based on the preset total dialysis duration, the current remaining dialysis duration is obtained; by combining the remaining task volume and the remaining dialysis duration, the current rigid demand rate is obtained.

[0007] Furthermore, the method for configuring the parameters of the hemodialysis device includes: When the hemodynamic load index changes abruptly, the system enters a safety maintenance mode, using the minimum of the current rigid demand rate and the preset safe flow rate limit as the target removal rate for the next moment. When not in safe hold mode, the target removal rate for the next moment is obtained based on the distribution of the hemodynamic load index and the rigid demand rate.

[0008] Furthermore, when not in the safety hold mode, if the hemodynamic load index is less than the preset comfort threshold, the current rigid demand rate is increased based on the difference between the hemodynamic load index and the preset comfort threshold to obtain the target removal rate at the next moment; If the hemodynamic load index is greater than or equal to the preset comfort threshold and less than the preset warning threshold, the minimum value of the current vascular refill rate and the rigid demand rate is taken as the target removal rate for the next moment. If the hemodynamic load index is greater than or equal to a preset warning threshold, the current vascular refill rate is reduced based on a preset recovery coefficient to obtain the target removal rate for the next moment.

[0009] Furthermore, after obtaining the target removal rate, it also includes: The target removal rate is adjusted by comparing the current real-time removal rate with the target removal rate at the next moment, based on a preset maximum flow rate change range.

[0010] Furthermore, the method for obtaining the vascular refill rate includes: Obtain the instantaneous rate of change of the current capacity decay fundamental wave, fuse the instantaneous rate of change and the corrected capacity response coefficient to obtain the current net capacity reduction rate; subtract the net capacity reduction rate from the real-time removal rate to obtain the vascular refill rate.

[0011] Furthermore, the method for obtaining the modified capacity response coefficient includes: The initial capacity response coefficient is corrected by mapping the pipe wall shrinkage ratio using a nonlinear correction factor, and the current corrected capacity response coefficient is obtained.

[0012] Furthermore, the method for obtaining the pipe wall shrinkage ratio includes: By comparing the root mean square value of the tube wall pulsation wave data within the current preset neighborhood window with the amplitude of the relaxed pulsation, the tube wall contraction ratio is obtained.

[0013] The present invention also proposes an intelligent parameter configuration system for a hemodialysis device, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement any of the steps of the intelligent parameter configuration method for a hemodialysis device.

[0014] The present invention has the following beneficial effects: This invention first obtains the initial capacity response coefficient and the relaxation-state pulsation amplitude to provide a fluid dynamics reference system. It then extracts the vessel wall contraction ratio, directly quantifying the degree of vascular contraction and hardening. Combined with the initial capacity response coefficient, a corrected capacity response coefficient is obtained, enabling subsequent parameter settings for the hemodialysis device to compensate for the impact of vascular compliance changes caused by vascular contraction on the capacity-refill mapping relationship. Furthermore, the vascular refill rate is obtained, eliminating interference from vascular contraction and reflecting the current, most realistic upper limit of physiological supply capacity, providing an important basis for configuring the hemodialysis device parameters. Further comparison of the current vascular refill rate and the real-time removal rate, combined with the distribution of the capacity decay fundamental wave, yields the hemodynamic load index, characterizing the patient's current comprehensive physiological stress level. Finally, the current rigid demand rate is obtained based on global task state variables. Based on the current hemodynamic load index, rigid demand rate, and vascular refill rate, the parameters of the hemodialysis device are configured, realizing a shift from hysteresis feedback to proactive adaptive control, improving dehydration efficiency while ensuring safety. This invention accurately obtains the true vascular refill rate by calculating the vascular contraction ratio in real time and correcting the volume response coefficient; then, by dynamically configuring ultrafiltration parameters in combination with hemodynamic load and task requirements, it effectively overcomes the problem of deviation in the estimation of refill rate by traditional linear models due to vascular contraction. Attached Figure Description

[0015] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 A flowchart illustrating a method for intelligent parameter configuration of a hemodialysis device according to an embodiment of the present invention; Figure 2 This is a flowchart illustrating a method for obtaining the target removal rate when not in a safe holding mode, as provided in an embodiment of the present invention. Detailed Implementation

[0017] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a parameter intelligent configuration method and system for a hemodialysis device according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0019] The following description, in conjunction with the accompanying drawings, details the specific scheme of the intelligent parameter configuration method and system for a hemodialysis device provided by the present invention.

[0020] Please see Figure 1 The diagram illustrates a flowchart of a method for intelligent parameter configuration of a hemodialysis device according to an embodiment of the present invention, specifically including: Step S1: Within the preset initial calibration time window, obtain the initial capacity response coefficient and the relaxation state pulsation amplitude.

[0021] In order to achieve accurate assessment of subsequent changes in vascular status, it is necessary to establish a hydrodynamic reference system based on the individual physiological characteristics of the patient and in a state of vascular relaxation. Therefore, within the preset initial calibration time window, the initial volume response coefficient and the relaxation state pulsation amplitude are obtained.

[0022] Preferably, in one embodiment of the present invention, the initial calibration time window is 20 minutes long, the data acquisition frequency is 1 Hz, and the observed values ​​of whole blood relative volume fed back by the sensor are collected. .

[0023] Since it takes time for the fluid circuit system of a hemodialysis device to establish a steady flow field, and there is strong mechanical pump pressure noise in the extracorporeal circulation pipeline, it is difficult to directly extract weak vascular pulsation features from the original signal. Therefore, constant flow rate excitation is first performed to build a stable observation environment, and adaptive notch filtering technology is used for noise reduction.

[0024] Within 0-2 minutes, the ultrafiltration pump of the control unit operates at a constant calibrated removal rate. Run. In this embodiment, Set as It remains unchanged within the initial calibration time window, forming a step input signal.

[0025] Using a cutoff frequency of The low-pass filter. This channel outputs a capacity that attenuates the fundamental frequency. It is used to filter out all high-frequency fluctuations and retain only the macroscopic change trajectory that reflects the total volume removal over time.

[0026] To accurately extract weak cardiac-originating vessel wall pulsations from the raw signal containing strong background noise (mainly periodic mechanical pulsations caused by peristaltic pumps), this embodiment employs an adaptive comb notch filter: First, the system reads the speed feedback of the ultrafiltration pump drive unit in real time to obtain the current main pump frequency. (Usually between 0.5Hz and 5Hz).

[0027] Next, construct a frequency domain targeting and its second harmonic ( ), third harmonic ( The notch filter deeply attenuates the energy near these specific frequency points (suppression ratio >40dB).

[0028] Finally, the signal after notch filtering is passed through a passband... The bandpass filter.

[0029] The output of this channel is the tube wall pulsation wave. This processing logic ensures that the extracted signal primarily contains cardiac vascular pulsation components within the heart rate frequency band that do not overlap with the mechanical pump frequency. To further improve the signal-to-noise ratio, it is also possible to combine the ECG signal or estimated heart rate for synchronous superposition and averaging, or use wavelet transform to extract energy characteristics in the 0.8–2.0 Hz frequency band as a substitute measure of pulsation amplitude, thereby solving the problem that conventional filtering cannot remove mechanical interference in the same frequency band.

[0030] To eliminate transient response errors during the initial startup of the filter, the system... During the first two minutes, only data buffering is performed; no parameter calculations are performed.

[0031] In the initial minutes after ultrafiltration initiation, the refilling of interstitial fluid into the blood vessels has not yet significantly begun, and the decrease in blood volume is primarily driven by the mechanical removal action of the ultrafiltration pump. Furthermore, in the early stages of treatment, the patient's blood vessels have not yet undergone compensatory contraction and are in a relaxed state conforming to linear Hooke's law. Therefore, this physical window is utilized for linear regression identification to obtain baseline parameters characterizing the initial mechanical properties of the patient's blood vessels.

[0032] Specifically, select a time window (Corresponding to minutes 2 to 5). The system reads the capacity-attenuated fundamental frequency within this window. The data sequence was analyzed, and a least-squares linear fit was performed to obtain the absolute value of the capacity descent slope. (unit: ).

[0033] Based on the principle of fluid volume conservation, and ignoring the slight recharge, the initial capacity response coefficient is calculated. This characterizes the sensitivity of the relative change in whole blood volume caused by the removal of a unit volume of blood in a relaxed state (i.e., the reciprocal of the vascular system's compliance), and its physical meaning is the reciprocal of the percentage decrease in relative volume corresponding to the removal of 1 mL of blood. To prevent calculation overflow caused by the slope approaching zero due to extremely rapid patient refill, this embodiment applies a safety threshold constraint to the denominator: ; in, The preset minimum effective slope threshold is set to [value] in this embodiment. (Corresponding to the maximum physiological compliance limit to prevent overflow of calculation parameters). The unit is This parameter quantifies how many milliliters of blood, removed during the initial vasodilation phase of treatment, would result in a 1% decrease in observed total blood volume. The system... At any time The value is locked and stored as a reference constant for dynamically calculating the correction coefficient in subsequent stages.

[0034] Meanwhile, considering that the tangential elastic modulus of the blood vessel wall directly determines its mechanical response amplitude to cardiac pressure waves, and given that patients are in a baseline relaxed state at the beginning of hemodialysis, their pulsation amplitude reflects the inherent elastic characteristics of the blood vessel wall, baseline pulsation feature extraction is performed to obtain the relaxation-state pulsation amplitude.

[0035] Specifically, selection and Same time window The system reads the noise-reduced pipe wall pulsation wave within this window. The data uses the root mean square (RMS) value of the data sequence as the amplitude of the relaxed-state pulsation. .

[0036] To ensure the validity of the benchmark value, the system will perform calculations... Perform threshold validation. If... This indicates that the signal-to-noise ratio is too low or the sensor has poor contact. The system will issue an alarm and request an inspection of the sensor coupling or recalibration, for example, by selecting a sensor of the same length to the reverse. The data is recalculated using the data in the new window, and can be recalibrated a maximum of three times, such as [300,480], [480,660], [660,840]; otherwise, the obtained data is... (unit: It is locked and stored.

[0037] It should be noted that, and For calibration data within the same window range, if Recalibrate, then It will also select the corresponding time range for recalibration; if the calibration is successful... and The system will determine that the calibration has failed and automatically downgrade to the basic safety maintenance mode.

[0038] To prevent calibration parameters from deviating from the normal physical range due to the patient's specific physiological state (such as extreme dehydration or overhydration), the self-consistency of the parameters needs to be verified using data from the end of the calibration phase before the calibration phase ends.

[0039] Specifically, select the calibration end window. (Corresponding to minutes 15 to 20). The system reads the contents of this window. Data, calculate its steady-state descent slope Utilize the locked Reverse calculate the current theoretical recharge rate: ; like Within a reasonable physiological range (e.g.) If the calibration is successful, the system will determine that the calibration was successful. The initial calibration phase officially ends at this point, and the system state is switched to real-time control mode, carrying the locked static constants. and Moving on to the next stage.

[0040] like If the parameters are not within a reasonable physiological range, the parameter self-consistency verification is deemed to have failed, and the system will automatically downgrade to the basic safety maintenance mode.

[0041] It should be noted that in other embodiments of the present invention, the implementer may adjust the data acquisition frequency, window setting method and empirical threshold setting as needed, which will not be elaborated further.

[0042] Step S2: Based on the fluctuation of the vessel wall pulsation wave data within the current preset neighborhood window, and combined with the relaxation state pulsation amplitude, obtain the vessel wall contraction ratio; fuse the current vessel wall contraction ratio and the initial volume response coefficient to obtain the corrected volume response coefficient; based on the instantaneous change of the volume decay fundamental wave, and combined with the corrected volume response coefficient, compare the real-time removal rate of the ultrafiltration pump to obtain the vessel refill rate.

[0043] Step S2 is executed after switching to real-time control mode and after the initial capacity response coefficient and relaxation state pulsation amplitude have passed self-consistency verification.

[0044] As dialysis progresses and effective circulating blood volume decreases, excitation of the sympathetic nervous system triggers vascular smooth muscle contraction, leading to increased tangential stiffness of the vessel wall. This stiffening effect directly reduces the amplitude of vessel wall pulsations caused by the cardiac cycle. If the model ignores this physical change, it will misjudge the vascular volume status. Therefore, based on the fluctuations of vessel wall pulsation wave data within the current preset neighborhood window, combined with the relaxation-state pulsation amplitude, the vessel wall contraction ratio is obtained. By quantifying the degree of pulsation attenuation, the contraction state of the blood vessel is characterized, directly quantifying the degree of vascular contraction stiffness. This provides crucial dynamic input parameters for subsequent correction of monitoring signal distortion caused by vascular contraction.

[0045] Preferably, in one embodiment of the present invention, the root mean square value of the vascular pulsation signal is a direct measure of its energy or average amplitude. The amplitude of the relaxed-state pulsation, measured when the blood vessels are in a relaxed state at the beginning of dialysis, establishes an individualized vascular elasticity benchmark for the patient. As dialysis progresses, the blood vessels harden due to compensatory contraction, reducing their compliance with cardiac cycle pressure waves, resulting in a decrease in the amplitude of the vessel wall pulsation.

[0046] By comparing the root mean square value of the tube wall pulsation wave data within the current preset neighborhood window The tube wall contraction ratio is obtained by comparing the amplitude of the relaxed-state pulsation. This allows for precise quantification of the dynamic changes in vascular wall tension. Simultaneously, to prevent sensor artifacts from causing numerical spikes, or to prevent abnormal vasodilation in the early stages of hypotensive shock leading to abnormal ratios, numerical clamping protection is applied to the vessel wall contraction ratio.

[0047] As an example, the current As molecules, the amplitude of the relaxed-state pulsation Adding the sum of preset positive parameters divided by zero as the denominator, the fractional ratio is clamped to a preset first interval, and the result is taken as the current pipe wall shrinkage ratio. .

[0048] The root mean square (RMS) value is used to represent the fluctuation of the tube wall pulsation wave data; the preset value of the positive parameter divided by zero is... The dimensions are the same as the parameters directly added together. In other fractions in this embodiment of the invention, this method can be used to prevent the denominator from being 0; the preset first interval is set as The lower limit of 0.1 is used to prevent the risk of aggressive control caused by parameter anomalies due to signal loss, while the upper limit of 1.5 is used to filter out non-physiological numerical spikes caused by motion artifacts. The smaller the value, the more severe the vasoconstriction; a value close to 1 indicates that the blood vessel is maintained in a baseline relaxed state; and a value greater than 1 may indicate that the blood vessel has undergone passive vasodilation.

[0049] When blood vessels constrict and harden, the removal of a unit volume of fluid will cause a change in the tube wall contraction ratio, and the initial capacity response coefficient will no longer be fully applicable. Therefore, by combining the current tube wall contraction ratio and the initial capacity response coefficient, a corrected capacity response coefficient can be obtained, so that the subsequent parameter settings of the hemodialysis device can compensate for the impact of changes in vascular compliance caused by vascular constriction on the capacity-refill mapping relationship.

[0050] Preferably, in one embodiment of the present invention, the change in vascular compliance conforms to the nonlinear stress-strain relationship commonly found in biological soft tissues. When a blood vessel begins to contract from a relaxed state, the initial slight contraction leads to a significant decrease in compliance (i.e., the vascular expandability); however, as the degree of contraction intensifies, the rate of decrease in compliance slows down. Therefore, a nonlinear correction factor is used to map the vessel wall contraction ratio to correct the initial volume response coefficient. Obtain the current corrected capacity response coefficient.

[0051] As an example, nonlinear correction factor These are approximate engineering parameters obtained based on historical data regression analysis. Their value range is set as follows: In this example, we use 0.5. To utilize the mapping results of the pipe wall shrinkage ratio using a nonlinear correction factor, and The product of these factors is used as the corrected capacity response coefficient. ; In order to prevent extreme spasmodic contraction of blood vessels (such as...) Under extreme conditions approaching 0.1, the corrected compliance coefficient is too small, leading to a non-physiologically inflated (i.e., misjudged as extremely strong refill capacity) calculated subsequently. The lower limit is set as The threshold setting of 0.2 is based on biomechanical characteristics. Even if the blood vessels undergo maximum physiological contraction, their residual compliance is usually not less than 20% of the relaxed state.

[0052] The instantaneous rate of change in blood volume readings is a mixed result, encompassing both the active removal action of the ultrafiltration pump and the passive refilling of tissue fluid into the blood vessels. The true refill status cannot be determined directly from this mixed signal. However, based on the principle of mass conservation of fluids, and using the instantaneous change in the fundamental frequency of volume decay, combined with a corrected volume response coefficient, the real-time removal rate of the ultrafiltration pump can be compared to obtain the vascular refill rate. This decouples the pure tissue fluid refill rate from the mixed signal. The vascular refill rate, free from the interference of vasoconstriction, reflects the patient's current, most realistic physiological supply capacity, providing crucial information for configuring parameters for hemodialysis devices.

[0053] In a preferred embodiment of the present invention, based on the law of conservation of fluid mass, in the extracorporeal circulation loop of a hemodialysis device, the net rate of change of intravascular blood volume is determined by two opposing flow rates: the "removal rate" by which the ultrafiltration pump removes fluid from the body, and the "refill rate" by which interstitial fluid flows back into the blood vessels. The instantaneous rate of change of the fundamental wave of volume decay is the slope of the decrease in observed blood volume, which directly reflects the actual physical change in intravascular volume, but its magnitude is significantly modulated by the current vascular compliance (i.e., the corrected volume response coefficient).

[0054] Based on this, the instantaneous rate of change of the current capacity decay fundamental wave is obtained, and the instantaneous rate of change and the corrected capacity response coefficient are fused to obtain the current net capacity reduction rate; the net capacity reduction rate is subtracted from the real-time removal rate to obtain the vascular refill rate.

[0055] As an example, to reduce differential noise, a five-point difference method is used to obtain the instantaneous rate of change of the current capacity attenuation fundamental frequency. The numerator is calculated by subtracting the capacity attenuation fundamental frequency from the capacity attenuation fundamental frequency four times prior, and the denominator is the sum of the values ​​from the four acquisition intervals. The product of this fraction and 60 is then used as the current capacity attenuation fundamental frequency. The multiplication by 60 is to convert the result to the rate of change per minute (% / min).

[0056] The absolute value of the instantaneous rate of change and the corrected volume response coefficient are multiplied and combined to obtain the net volume reduction rate, which accurately calculates the actual physical reduction rate of the current intravascular volume. The difference between the current real-time removal rate of the ultrafiltration pump and the net volume reduction rate is taken as the vascular refill rate. .

[0057] Among them, as ultrafiltration proceeds, the relative volume of whole blood... The overall trend is downward (i.e., blood concentration), therefore Typically negative; there is a non-negativity constraint on vascular refill rate, i.e., it is truncated at 0, and values ​​less than 0 are directly assigned 0. The value represents the true physiological supply capacity indicator after eliminating the physical interference of vasoconstriction.

[0058] Step S3: Compare the current vascular refill rate and the real-time removal rate, and obtain the hemodynamic load index by combining the distribution of the volume decay fundamental wave; obtain the current rigid demand rate based on the global task state variables; configure the parameters of the hemodialysis device according to the current hemodynamic load index, rigid demand rate and vascular refill rate.

[0059] The vascular refill rate represents the maximum supply capacity of the physiological system to replenish blood volume, while the real-time removal rate represents the current demand for blood volume consumed by the hemodialysis device. The comparison between the two directly reflects the instantaneous supply-demand balance of the circulatory system. The distribution of the volume decay fundamental wave reflects the depth of volume consumption caused by cumulative dehydration load. Combining instantaneous supply-demand pressure with the risk of cumulative volume consumption for fusion assessment generates a comprehensive hemodynamic load index. This index overcomes the limitations of single-dimensional assessment, quantitatively characterizing the patient's current comprehensive physiological stress level, providing a graded and precise basis for risk assessment from "relaxed" to "urgent," and enabling early and comprehensive warning of hypotension risk.

[0060] Preferably, in one embodiment of the present invention, the current vascular refill rate and the real-time removal rate are first compared to obtain the dynamic supply and demand factor; Furthermore, based on the degree of approximation of the fundamental wave of capacity decay to the preset capacity threshold, the trajectory of blood volume observation over time and its degree of approximation relative to the preset safety threshold are analyzed to obtain the static inventory factor, which characterizes the depth of capacity consumption caused by cumulative dehydration load. Finally, by combining the current dynamic supply and demand factors and static inventory factors, the hemodynamic load index is obtained.

[0061] As an example, the current real-time removal rate is used as the numerator, the sum of the vascular refill rate and the preset baseline safety margin is used as the denominator, and the ratio of the fractions is used as the current dynamic supply and demand factor; in this example, the preset baseline safety margin is... This is used to prevent mathematical singularities where the denominator is zero.

[0062] Further, the capacity attenuation fundamental wave at time k=120 after the data buffering ends. Based on the fundamental frequency with capacity attenuation, divide by To perform maximum value normalization (setting it to 1 if it exceeds 1), the normalized capacity attenuation fundamental frequency is obtained. Set a preset safety threshold. for 85%, meaning a 15% reduction is allowed; Calculate the approximation factor , , This represents the preset minimum safe distance, which is 0.5% in this example. It shows how close the fundamental frequency of capacity attenuation is to the preset capacity threshold; Finally, the preset distance scaling factor is applied. Divide by The square of the quotient is used as the current static inventory factor. The product of the static inventory factor and the dynamic supply and demand factor is used as the current hemodynamic load index. . In this example, it is set to 5%.

[0063] in, The setting is to prevent the denominator from becoming negative or zero when the actual blood volume falls below the safety threshold, ensuring that the risk factor calculated in the extremely dangerous zone is still positive and at a saturated high level; the design of the square operation is to construct a non-linear risk gradient, so that when the blood volume approaches the critical line, the risk reading is exponentially amplified, thereby triggering the system's strong intervention at the edge of danger. It serves as a scaling benchmark to map dimensionless distance ratios to a range of risk indices that the system can identify.

[0064] It should be noted that the parameter settings in the hemodynamic load index calculation process are based on the empirical calibration results determined by statistical analysis and control simulation experiments of a large number of clinical dialysis case data. In other embodiments of the present invention, implementers can adjust them according to the specific implementation scenario, which will not be elaborated further.

[0065] The hemodialysis process must achieve the preset dehydration target under safe conditions. Analyzing the global task state variables can be done purely from a task perspective, obtaining the current rigid demand rate. The hemodynamic load index provides a quantitative assessment of the current physiological safety status, and the vascular refill rate marks the insurmountable upper limit of physiological supply capacity. Therefore, based on the current hemodynamic load index, rigid demand rate, and vascular refill rate, the parameters of the hemodialysis device are configured. Based on the configuration method of multi-source information fusion, a fundamental shift from passive safety monitoring to active physiological adaptive control is achieved. Ultimately, while ensuring patient safety, the reliability and efficiency of the blood dehydration process are significantly improved.

[0066] Preferably, in one embodiment of the present invention, the remaining workload is first obtained based on a preset total ultrafiltration target and the current and previous real-time removal rates. That is, by accumulating and integrating the real-time removal rate at each current and historical moment (existing technology), the amount of ultrafiltration is obtained; the difference between the preset total ultrafiltration target amount and the amount of ultrafiltration is taken as the remaining task amount.

[0067] Then, based on the preset total dialysis time, the current remaining dialysis time is obtained: the preset total dialysis time is subtracted from the current treatment time, and the difference is taken as the remaining dialysis time (in minutes). Next, by combining the remaining workload and remaining dialysis time, the current rigid demand rate is obtained: the quotient of the current remaining workload divided by the remaining dialysis time is taken as the current rigid demand rate. .

[0068] The preset total ultrafiltration target and preset total dialysis duration are parameters set in the patient's prescription and will not be further specified here. The current treatment duration can be obtained based on the current k-index and acquisition frequency, which is existing technology and will not be elaborated further. To prevent the remaining dialysis duration from being too short when treatment is about to end, leading to calculation overflow or rate mutation, a locking logic is set: if the current remaining dialysis duration is less than 5 minutes, it will not be updated. It directly uses and locks the value from the previous moment.

[0069] Furthermore, after entering real-time control mode (i.e., after passing the initial calibration time window), the hemodynamic load index is taken into account. When a mutation occurs, it indicates a step mutation that exceeds the gradual range of normal physiological regulation. This may reflect abnormal data or a high-risk pathological event. At this point, it is determined to enter a safety maintenance mode, reducing the current rigid demand rate. With preset safe flow rate limit The minimum value is used as the target removal rate for the next time step; As an example, The determination condition is: existence. The relative change rate of the hemodynamic load index within an adjacent minute should exceed 50%.

[0070] When not in safe hold mode, the target removal rate for the next time step is obtained based on the distribution of the hemodynamic load index and the rigid demand rate. Please refer to [link to relevant documentation]. Figure 2 The diagram illustrates a flowchart of a method for obtaining the target removal rate when not in a safety hold mode, according to an embodiment of the present invention, specifically including: Step S301: If the hemodynamic load index is less than the preset comfort threshold, increase the current rigid demand rate based on the difference between the hemodynamic load index and the preset comfort threshold, and obtain the target removal rate at the next moment.

[0071] If the hemodynamic load index is less than the preset comfort threshold, it indicates that the blood vessels are in a low-tension relaxation state and the tissue refill rate is greater than the current removal rate. This means that the physiological system has the capacity to handle additional dehydration load. At this time, the system actively increases the current rigid demand rate and removes water in advance using the early safety margin, thereby reducing the dehydration pressure in the later stages of hemodialysis.

[0072] As an example, the difference between the preset comfort threshold and the hemodynamic load index is multiplied by the preset forward aggressive coefficient, and then a constant 1 is added to it as the amplification coefficient. The product of the amplification coefficient and the current rigid demand rate is used as the initial removal rate at the next moment.

[0073] Meanwhile, to prevent excessively aggressive approaches from causing momentary imbalances, a target removal rate is obtained by constraining the initial removal rate: the target removal rate is at most 1.2 times the current vascular refill rate and does not exceed the maximum physical flow rate allowed by the dialysis machine hardware.

[0074] The preset comfort threshold is 0.8, and the preset forward aggressiveness coefficient is 0.5.

[0075] Step S302: If the hemodynamic load index is greater than or equal to the preset comfort threshold and less than the preset warning threshold, take the minimum value of the current vascular refill rate and rigid demand rate as the target removal rate for the next moment.

[0076] If the hemodynamic load index is greater than or equal to the preset comfort threshold and less than the preset warning threshold, it indicates that the supply and demand relationship is in a tight balance. The control objective is primarily to complete the rigid task, but it must be strictly limited by the physical boundaries of physiological supply capacity.

[0077] Therefore, the minimum of the current vascular refill rate and the rigid demand rate is taken as the target removal rate for the next moment.

[0078] Step S303: If the hemodynamic load index is greater than or equal to the preset warning threshold, reduce the current vascular refill rate based on the preset recovery coefficient to obtain the target removal rate at the next moment.

[0079] If the hemodynamic load index is greater than or equal to the preset warning threshold, it indicates that the blood vessels are in a state of high-tension contraction, or that the blood volume reserve is approaching the critical line, suggesting that the physiological regulatory capacity is nearing its limit. The system abandons rigid task constraints and, based on the preset recovery coefficient, reduces the current vascular refill rate to prioritize safety.

[0080] As an example, the preset recovery coefficient is set to 0.8, and the product of the preset recovery coefficient and the current vascular refill rate is used as the target removal rate for the next moment.

[0081] Because a sudden, abrupt change in flow rate can induce a spasmodic reflex in vascular smooth muscle, after obtaining the target removal rate, the following steps are also taken: The target removal rate is adjusted by comparing the current real-time removal rate with the target removal rate at the next moment, based on the preset maximum flow rate change range.

[0082] As an example, the preset maximum flow rate variation is This involves comparing the current real-time removal rate with the target removal rate to determine the direction of adjustment (increase or decrease), and then limiting the maximum adjustment range to [value missing]. .

[0083] It should be noted that the preset comfort threshold, preset warning threshold, forward aggressive coefficient, recovery coefficient, upper limit multiple of refill rate, and maximum flow rate change range are all derived from historical treatment data and control simulation calibration under typical clinical scenarios. In other embodiments of the present invention, the implementer may adjust them according to clinical needs or equipment configuration. In another embodiment of the present invention, a condition for determining the safety maintenance mode can be added, such as detecting the root mean square value of the tube wall pulsation wave data within the current preset neighborhood window. When the value is less than 0.005%, it indicates that the signal is lost or extremely weak, and in this case, you can also enter the safety hold mode.

[0084] An embodiment of the present invention also provides a parameter intelligent configuration system for a hemodialysis device. The system includes a memory, a processor, and a computer program. The memory is used to store the corresponding computer program, and the processor is used to run the corresponding computer program. When the computer program runs in the processor, it can implement the parameter intelligent configuration method for a hemodialysis device described in steps S1-S3.

[0085] In summary, to address the technical problem of inaccurate ultrafiltration control caused by neglecting compliance changes resulting from vasoconstriction in existing technologies, this invention provides an intelligent parameter configuration method and system for hemodialysis devices. This invention first obtains the initial volume response coefficient and relaxation-state pulsation amplitude; further, it extracts the vessel wall contraction ratio and, combined with the initial volume response coefficient, obtains a corrected volume response coefficient; further, based on the instantaneous change of the volume decay fundamental wave and combined with the corrected volume response coefficient, it compares the real-time removal rate of the ultrafiltration pump to obtain the vascular refill rate; further, it compares the current vascular refill rate and the real-time removal rate, and, combined with the distribution of the volume decay fundamental wave, obtains the hemodynamic load index; finally, it obtains the current rigid demand rate based on global task state variables; and, based on the current hemodynamic load index, rigid demand rate, and vascular refill rate, configures the parameters of the hemodialysis device. This effectively overcomes the problem of inaccurate refill rate estimation in traditional linear models caused by vasoconstriction, achieving a shift from hysteresis feedback to proactive adaptive control, improving dehydration efficiency while ensuring safety. This invention accurately obtains the true vascular refill rate by calculating the vascular contraction ratio in real time and correcting the volume response coefficient; then, by dynamically configuring ultrafiltration parameters in combination with hemodynamic load and task requirements, it effectively overcomes the problem of deviation in the estimation of refill rate by traditional linear models due to vascular contraction.

[0086] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0087] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for intelligent parameter configuration of a hemodialysis device, characterized in that, The method includes: Within the preset initial calibration time window, the initial capacity response coefficient and relaxation-state pulsation amplitude are obtained; Based on the fluctuations of the vessel wall pulsation wave data within the current preset neighborhood window, and combined with the relaxation state pulsation amplitude, the vessel wall contraction ratio is obtained; the current vessel wall contraction ratio and the initial capacity response coefficient are fused to obtain the corrected capacity response coefficient; based on the instantaneous change of the capacity decay fundamental wave, and combined with the corrected capacity response coefficient, the real-time removal rate of the ultrafiltration pump is compared to obtain the vessel refill rate. By comparing the current vascular refill rate and the real-time removal rate, and combining the distribution of the volume decay fundamental wave, the hemodynamic load index is obtained; the current rigid demand rate is obtained based on the global task state variables; and the parameters of the hemodialysis device are configured according to the current hemodynamic load index, the rigid demand rate, and the vascular refill rate.

2. The intelligent parameter configuration method for a hemodialysis device according to claim 1, characterized in that, The method for obtaining the hemodynamic load index includes: By comparing the current vascular refill rate and the real-time removal rate, a dynamic supply-demand factor is obtained; based on the degree to which the fundamental value of capacity decay approaches a preset capacity threshold, a static inventory factor is obtained; and by fusing the current dynamic supply-demand factor and the static inventory factor, a hemodynamic load index is obtained.

3. The intelligent parameter configuration method for a hemodialysis device according to claim 1, characterized in that, The method for obtaining the rigid demand rate includes: Based on the preset total ultrafiltration target and the current and previous real-time removal rates, the remaining task volume is obtained; based on the preset total dialysis duration, the current remaining dialysis duration is obtained; by combining the remaining task volume and the remaining dialysis duration, the current rigid demand rate is obtained.

4. The intelligent parameter configuration method for a hemodialysis device according to claim 1, characterized in that, The method for configuring the parameters of the hemodialysis device includes: When the hemodynamic load index changes abruptly, the system enters a safety maintenance mode, using the minimum of the current rigid demand rate and the preset safe flow rate limit as the target removal rate for the next moment. When not in safe hold mode, the target removal rate for the next moment is obtained based on the distribution of the hemodynamic load index and the rigid demand rate.

5. The intelligent parameter configuration method for a hemodialysis device according to claim 4, characterized in that, When not in safety hold mode, if the hemodynamic load index is less than the preset comfort threshold, the current rigid demand rate is increased based on the difference between the hemodynamic load index and the preset comfort threshold to obtain the target removal rate for the next moment. If the hemodynamic load index is greater than or equal to the preset comfort threshold and less than the preset warning threshold, the minimum value of the current vascular refill rate and the rigid demand rate is taken as the target removal rate for the next moment. If the hemodynamic load index is greater than or equal to a preset warning threshold, the current vascular refill rate is reduced based on a preset recovery coefficient to obtain the target removal rate for the next moment.

6. The intelligent parameter configuration method for a hemodialysis device according to claim 4 or 5, characterized in that, After obtaining the target removal rate, the following is also included: The target removal rate is adjusted by comparing the current real-time removal rate with the target removal rate at the next moment, based on a preset maximum flow rate change range.

7. The intelligent parameter configuration method for a hemodialysis device according to claim 1, characterized in that, The method for obtaining the vascular refill rate includes: Obtain the instantaneous rate of change of the current capacity decay fundamental wave, fuse the instantaneous rate of change and the corrected capacity response coefficient to obtain the current net capacity reduction rate; subtract the net capacity reduction rate from the real-time removal rate to obtain the vascular refill rate.

8. The intelligent parameter configuration method for a hemodialysis device according to claim 1, characterized in that, The method for obtaining the corrected capacity response coefficient includes: The initial capacity response coefficient is corrected by mapping the pipe wall shrinkage ratio using a nonlinear correction factor, and the current corrected capacity response coefficient is obtained.

9. The intelligent parameter configuration method for a hemodialysis device according to claim 1, characterized in that, The method for obtaining the pipe wall shrinkage ratio includes: By comparing the root mean square value of the tube wall pulsation wave data within the current preset neighborhood window with the amplitude of the relaxed pulsation, the tube wall contraction ratio is obtained.

10. A parameter intelligent configuration system for a hemodialysis device, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the intelligent parameter configuration method for a hemodialysis device as described in any one of claims 1 to 9.

Images