Search

CN-122005986-A - Intelligent hemodialysis system based on sodium removal ratio and control method thereof

CN122005986ACN 122005986 ACN122005986 ACN 122005986ACN-122005986-A

Abstract

The invention discloses an intelligent hemodialysis system based on a sodium removal ratio (R_Na) and a control method thereof. The system integrates five layers of functions to form a complete closed loop, namely (1) a sensing layer, based on a dialysate conductivity difference value, a correction Gibbs-Donnan formula (C_eq= [ Na ] d/sigma, sigma=0.97) and Kalman filtering are applied, plasma sodium concentration is reversely pushed in real time, (2) a calculating layer, namely, R_Na=Na_actual/Na_target is calculated in real time, (3) a monitoring layer, namely, R_Na+Kt/V double-index instrument panel+three-level alarm, (4) a control layer, namely, a double-chamber model, namely, na d regulation curve is calculated, R_Na closed loop regulation and three safety rules are applied, and (5) a feedback layer, namely, cross-treatment course trend analysis and model automatic optimization are performed. The system is compatible with dialysis machines of various brands and can be integrated as an additional module.

Inventors

  • LIU ZIDONG

Assignees

  • 济南涧水科技服务有限公司

Dates

Publication Date
20260512
Application Date
20260325

Claims (15)

  1. 1. The control method of the intelligent hemodialysis system based on the sodium removal ratio is characterized by comprising the following steps of: And S1, estimating the plasma sodium concentration of the sensing layer in real time. During dialysis, measuring the inlet conductivity kappa_in and the outlet conductivity kappa_out of the dialysis fluid in real time, converting conductivity values into sodium concentration values by utilizing a pre-calibrated conductivity-sodium concentration conversion relation, calculating the inlet sodium concentration [ Na ] d_in and the outlet sodium concentration [ Na ] d_out of the dialysis fluid, calculating the sodium mass transfer quantity delta Na=Q_d× ([ Na ] d_out- [ Na ] d_in) per unit time according to the flow rate Q_d of the dialysis fluid, applying a corrected Gibbs-Donnan balance formula C_eq= [ Na ] d_out/sigma (sigma=0.97 is Gibbs-Donnan factor), calculating the dispersion balance concentration, and reversely pushing the current plasma sodium concentration [ Na ] pw ] x [ Na ] w to the serum sodium concentration [ Na ] s=0.93 x [ Na ] pw ] by combining the sodium mass balance formula delta Na=K_d× (C_Na ] pw) +Q_uf x [ Na ] pw; And S2, calculating the layer-sodium removal ratio R_Na in real time. Obtaining the weight W_pre of a patient before dialysis, the weight W_dry and the concentration [ Na ] s_0 of serum sodium before dialysis, calculating a target sodium removal amount Na_target= (W_pre-W_dry) x [ Na ] s_0/0.93, calculating an accumulated actual sodium removal amount Na_actual based on the time integral of the sodium mass transfer amount of the step S1, calculating a sodium removal ratio R_Na=Na_actual/Na_target in real time; And step S3, a monitoring layer, namely double-index display and alarm. Simultaneously displaying two indexes of R_Na and Kt/V on a dialysis equipment interface, wherein R_Na reflects sodium balance sufficiency, and Kt/V reflects urea clearance sufficiency; And S4, dynamically regulating and controlling the concentration of the sodium in the control layer-dialysate. In the dialysis process, according to the real-time R_Na value obtained in the step S2, dynamically correcting the regulation curve, namely reducing [ Na ] d 1-3 mmol/L to increase sodium discharge when R_Na is less than 0.9, and increasing [ Na ] d 1-3 mmol/L to reduce sodium discharge when R_Na is more than 1.1, thereby forming closed-loop control driving R_Na to approach 1.0; And S5, a feedback layer, namely evaluating and learning after dialysis. And calculating an R_Na final value and a Kt/V final value at the end of dialysis, analyzing the deviation and deviation sources of the actual sodium removal amount and the target value, and feeding back deviation information to be used for the initial parameter optimization and model calibration of the next dialysis.
  2. 2. The method according to claim 1, wherein the step S1 further comprises a non-sodium ion correction step of measuring or presetting the concentration of other main ions (potassium, calcium, magnesium, bicarbonate) in the dialysis liquid, calculating a non-sodium ion conductivity contribution value K_non-Na according to the contribution coefficient of each ion to the total conductivity, subtracting the non-sodium contribution from the total conductivity to obtain sodium home conductivity K_Na=K_total-K_non-Na, and using K_Na instead of K_total to calculate the sodium concentration to improve the estimation accuracy.
  3. 3. The method of claim 1, wherein the step S1 further comprises a signal filtering step of performing digital filtering processing on the [ Na ] pw original estimated value obtained by the back-extrapolation by using a Kalman filter, wherein the state variable of the Kalman filter is [ Na ] pw, the process model is based on a double-chamber sodium dynamics equation (including disturbance reference ICF exchange models j_ic=K_ic× [ (C_e-C_i) - (C_e0-C_i0) ]), and the observation model is based on a conductivity-sodium concentration conversion relation, so that the estimated value after filtering is compatible with measurement instantaneity and sodium dynamics physical constraint.
  4. 4. The method according to claim 1, wherein the step S1 further comprises an on-line verification and automatic calibration step of comparing the laboratory test value [ Na ] s_lab with the estimated value [ Na ] s_est at the same time when blood samples are collected during dialysis and the estimated value [ Na ] s_est is obtained, calculating the deviation delta= [ Na ] s_lab- [ Na ] s_est, and automatically correcting the conductivity-sodium concentration transfer function parameter and/or the estimated value of the diffusion clearance K_d when |delta| > 2 mmol/L, and storing the calibration result in a patient personal parameter file for subsequent dialysis.
  5. 5. The method of claim 1, wherein the dual index display in the step S3 uses a parallel instrument panel interface, the left side displays the real-time r_na value and the trend curve thereof with time, the right side displays the real-time Kt/V value and the trend curve thereof, and the two indexes respectively use independent target area identifiers, wherein the target area of r_na is [0.9, 1.1], the target area of Kt/V is not less than 1.2 (HD) or not less than 1.0 (on-line HDF), and the indexes are green when being in the target area, and yellow or red when being deviated.
  6. 6. The method according to claim 1, wherein the hierarchical alert in step S3 employs a three-level mechanism: Yellow warning that R_Na deviates slightly (R_Na is more than or equal to 0.8 and less than or equal to 0.9 or 1.1 and less than or equal to 1.2), and only a prompt is displayed on the machine; Orange alarm, wherein R_Na deviates moderately (R_Na is more than or equal to 0.7 and less than or equal to 0.8 or R_Na is more than or equal to 1.2 and less than or equal to 1.3), and the local machine displays and informs a responsible nurse; Red alarm-serious deviation of R_Na (R_Na < 0.7 or R_Na > 1.3), simultaneous alarm at the local, central monitoring station and doctor mobile terminal, and automatic triggering of the safety protection mode to lock [ Na ] d to the same level as serum sodium.
  7. 7. The method of claim 1, wherein the dialysate sodium concentration regulation curve in step S4 is calculated from a dual-chamber sodium kinetic model rather than being directly determined from a pre-determined empirical model, and the staging strategy is: in the first stage (first 40% -50% of dialysis time), na d is set to be close to or slightly higher than the serum sodium concentration of a patient, so that the sodium diffusion rate is reduced, the osmotic pressure of blood plasma is maintained stable, and early hypotension is avoided; In the second stage (the later 50-60% of dialysis time), na d is set to be lower than serum sodium concentration by 2-5 mmol/L, the gradient of sodium diffusion and sodium discharge is increased, and sodium discharge is accelerated to ensure that the final value of R_Na approaches 1.0.
  8. 8. The method according to claim 1, wherein the dynamic correction in step S4 further comprises the following rules: Blood pressure protection rules, namely when the systolic blood pressure of a patient is reduced by more than a preset threshold (20 mmHg), increasing the current [ Na ] d by 2-4 mmol/L and reducing the ultrafiltration rate, and recalculating a subsequent curve by a sodium kinetic model; The sodium clearance progress rule is that when the accumulated sodium clearance is lower than 80% of the target progress, the [ Na ] d is reduced by 1-2 mmol/L to accelerate sodium discharge, provided that the blood pressure is stable and the predicted serum sodium is not lower than the safety lower limit; serum sodium protection rules when the model predicts that serum sodium will be below the lower safety limit (135 mmol/L), the [ Na ] d is increased to reduce diffuse sodium excretion, and ultrafiltration volume is increased to compensate for convective sodium excretion if necessary.
  9. 9. The method according to claim 1, wherein the dialysate sodium concentration in step S4 is always locked within a safety range [130, 150] mmol/L, and is automatically locked to the nearest safety margin when the dynamically modified sodium concentration exceeds the range, the frequency of the dynamic modification being once every 5-15 minutes.
  10. 10. The method of claim 1, wherein step S5 further comprises cross-session trend management: recording R_Na final value and Kt/V final value at the end of each dialysis; Analyzing the R_Na trend of continuous multiple dialysis in a time sequence manner, and generating prescription adjustment advice when the R_Na final value deviates from the target range more than 3 times continuously; and generating a sodium balance track report of the week/month dimension, wherein the sodium balance track report comprises an R_Na mean value, a standard deviation, a standard reaching rate and a contrast analysis of the Kt/V standard reaching rate.
  11. 11. The method of claim 1, further comprising the step of normalizing the data communication: Packaging and outputting data including dialysis date and time, patient identification, weight before and after dialysis, serum sodium concentration before and after dialysis (containing real-time estimated value sequence), na_target, na_actual, R_Na final value, kt/V final value, dialysate sodium concentration execution curve and alarm event record by adopting an HL7 FHIR or IHE dialysis extension protocol; Transmitting the message to the HIS/EMR/DIMS through a network interface to realize the whole process traceability; the centralized monitoring of the R_Na of multiple patients is supported, and the real-time state of each dialysis machine R_Na is displayed by color coding at a central workstation.
  12. 12. The method of claim 1, wherein the method is compatible with the conductivity monitoring system of existing dialysis machines, including Gan Buluo/baud Diascan system, fei Senyou s OCM system, and other dialysis devices equipped with dialysate inlet and outlet conductivity sensors, wherein the compatibility is achieved by reading the signal output of the existing conductivity sensors without additional installation of conductivity sensors, wherein the method is applicable to Hemodialysis (HD), hemodiafiltration (HDF), and on-line hemodiafiltration (online-HDF), wherein the convective sodium removal in the sodium kinetic model also includes additional convective removal by the substitution fluid for the HDF and online-HDF modes.
  13. 13. An intelligent hemodialysis system based on a sodium removal rate, comprising: The plasma sodium concentration real-time estimation module comprises a dialysate inlet and outlet conductivity signal acquisition unit (or a signal interface with an existing dialysis machine conductivity sensor), a signal filtering unit (a Kalman filter) and a Donnan correction calculation unit (with built-in C_eq= [ Na ] d/sigma, sigma=0.97) which are used for back-pushing the plasma sodium concentration in real time from the conductivity difference value and converting the plasma sodium concentration into serum sodium concentration; The R_Na calculation module is used for calculating Na_target= (W_pre-W_dry) × [ Na ] s/0.93 according to patient parameters, calculating Na_actual based on the sodium mass transfer quantity accumulation value, and further calculating R_Na=Na_actual/Na_target in real time; The double-index display unit is used for simultaneously displaying R_Na and Kt/V real-time values and trend curves, and marking the standard-reaching state by using color codes; The dynamic regulation controller is internally provided with a double-chamber sodium kinetic model and is used for calculating a dialysis whole-process [ Na ] d regulation curve, and performing closed-loop regulation according to real-time R_Na feedback to drive R_Na to approach 1.0; The Na d regulation execution unit comprises a sodium concentration sensor and a concentrated solution proportioning regulation mechanism and is used for accurately regulating the concentration of sodium in the dialysate according to the instruction of the controller; the alarm module is used for generating three-level alarm signals when R_Na deviates from the target range; the data communication module is used for transmitting the system data to an external information system in a standardized format; and the feedback learning module is used for storing the execution record of each dialysis and the sodium removal result and automatically optimizing the model parameters.
  14. 14. The system according to claim 13, characterized in that the system is integrated as an add-on module with existing dialysis machines through standard communication interfaces (RS-232, USB, ethernet or wireless bluetooth) that support adaptation to different brands of dialysis machines (Fei Senyou s, bronzes, biot/Gan Buluo, nipro, etc.) and connection to AI Agent intelligent dialysis prescription optimization systems without the need to replace the whole dialysis equipment.
  15. 15. The system of claim 13, further comprising a dialysis report generation module for automatically generating a standardized adequacy assessment report after each dialysis session is completed, the report comprising: (a) R_Na final value and Kt/V final value of the dialysis and standard reaching judgment are carried out; (b) Time profile of r_na, kt/V and serum sodium concentration estimates during dialysis; (c) Dialysate sodium concentration regulation curve (planned versus actual performed value); (d) Comparing the accumulated sodium removal with the target sodium removal, and dispersing/separating item details; (e) R_na trend graph and sodium balance trace analysis of the last 10 dialyzed; (f) Prescription adjustment advice (if any).

Description

Intelligent hemodialysis system based on sodium removal ratio and control method thereof Technical Field The invention relates to the technical field of blood purification equipment and intelligent medical monitoring, in particular to an intelligent hemodialysis system based on a sodium removal ratio (Sodium Removal Ratio, R_Na) and a control method thereof, which integrate three functions of real-time estimation of plasma sodium concentration, on-line calculation and double-index monitoring of R_Na and dynamic closed-loop regulation and control of dialysate sodium concentration. Background Hemodialysis is the primary kidney replacement therapy for End Stage Renal Disease (ESRD) patients to sustain life. The evaluation of the adequacy of dialysis has long been based on the urea removal index Kt/V as a core criterion. Kt/V reflects the efficiency of small molecule solute (urea) removal, but does not reflect the sodium equilibrium, another critical therapeutic dimension in the dialysis process. Sodium is the main cation of the extracellular fluid, which determines the osmotic pressure and capacity of the extracellular fluid. Abnormal sodium balance in dialysis patients (insufficient sodium excretion resulting in volume overload, or excessive sodium excretion resulting in hypotension and muscle cramping) is a key factor affecting the long-term prognosis of patients. However, the existing dialysis equipment only takes Kt/V as an evaluation index of dialysis sufficiency, and lacks the capability of quantitatively monitoring the sodium removal effect. The sodium removal ratio r_na=na_actual/na_target proposed by the inventors is a completely new dialysis adequacy assessment indicator. Where na_target= (w_pre-w_dry) × [ Na ] s/0.93 is the target sodium clearance, na_actual is the actual sodium clearance calculated based on the sodium kinetic model. R_na=1.0 indicates that sodium clearance just meets the standard, r_na < 1.0 indicates insufficient sodium removal (high sodium dialysis), and r_na > 1.0 indicates excessive sodium removal (low sodium dialysis). However, to achieve real-time monitoring and closed-loop control of r_na on dialysis equipment, three technical challenges are faced: (1) how does the sense layer acquire plasma sodium concentration in real time independent of blood sampling? is (3) control layer-how is the "sodium curve" function of an existing dialysis machine to employ a fixed mode according to R _ Na real-time feedback to dynamically regulate dialysate sodium concentration? without being based on a sodium kinetic model, individualization is not possible. Existing dialysis machines (e.g., gan Buluo/baud Diascan system, fei Senyou s OCM system) have been equipped with dialysate inlet and outlet conductivity sensors, primarily for calculating the amount of ion dialysis to estimate Kt/V. Since sodium ions account for more than 95% of the total cations of the dialysate, conductivity is highly correlated with sodium concentration. However, the key disadvantage of the prior art method is that the conventional Gibbs-Donnan formula uses C_eq=σ× [ Na ] d (σ=0.97), and the correct formula is C_eq= [ Na ] d/σ, resulting in a systematic error of about 6%. Therefore, there is a need for an intelligent dialysis system that integrates real-time plasma sodium estimation, on-line R_Na monitoring and dynamic closed-loop [ Na ] d regulation. Disclosure of Invention The invention aims to solve the technical problems of providing an intelligent hemodialysis system based on a sodium removal ratio R_Na and a control method thereof, which integrate three functions of real-time estimation (sensing layer), real-time calculation and double-index monitoring (monitoring layer) of the concentration of plasma sodium, and dynamic closed-loop regulation (control layer) of the concentration of dialysate sodium to form a complete closed loop of sensing, calculating, monitoring, controlling and feedback. In order to achieve the above purpose, the invention adopts the following technical scheme: 1. Sensing layer-plasma sodium concentration real-time estimation And S1, collecting the conductivity in real time. The dialysate inlet conductivity κ_in and outlet conductivity κ_out are measured in real time at a frequency not less than once per minute. The dialysis machine can be used as a self-charged conductivity sensor (such as a Diascan system and an OCM system), and can be externally connected with a high-precision conductivity sensor. Step S2, conductivity-sodium concentration conversion. The conductivity value is converted to a sodium concentration value using a pre-calibrated conversion function f (κ). Optionally, the conductivity contribution of non-sodium ions is corrected. And S3, calculating sodium mass transfer quantity. Δna=q_d× ([ Na ] d_out- [ Na ] d_in). And S4, correcting the Donnan balance and the back-pushing of the sodium plasma. C_eq= [ Na ] d_out/σ (σ=0.97), and [ Na ] pw is back-pushed in combination with the m