Search

CN-121999963-A - Antibacterial drug administration process data monitoring and management system

CN121999963ACN 121999963 ACN121999963 ACN 121999963ACN-121999963-A

Abstract

The invention discloses a data monitoring and management system for an antibacterial drug administration process, in particular relates to the field of medical information data processing, and aims to solve the problem that drug resistance induction risks are difficult to dynamically quantify and manage in an antibacterial treatment process. The method comprises the steps of firstly, carrying out numerical fitting and metabolic dynamics analysis on collected time sequence monitoring data to construct an individual metabolic feature set and a bacterial instantaneous proliferation model, then, calculating a theoretical bacteriostasis rate based on a pharmacodynamic function, carrying out vector difference operation with a bacterial replication speed to generate a net inhibition gradient value, then, projecting the gradient value to a drug resistance risk interval, calculating a traversing retention time through time integration, and finally, carrying out parameter reverse iterative optimization taking the minimized retention time as a constraint to generate a drug administration parameter adjustment instruction. The invention constructs a digital closed loop for the drug-resistant bacteria game, and effectively reduces the risk of drug-resistant mutation enrichment.

Inventors

  • SU DAN

Assignees

  • 常州市第二人民医院
  • 医顺通信息科技(江苏)有限公司

Dates

Publication Date
20260508
Application Date
20260130

Claims (9)

  1. 1. The antibacterial drug administration process data monitoring and management system is characterized by comprising the following modules: The multidimensional feature mapping module is used for acquiring serum drug concentration monitoring values and inflammatory marker values according to time sequences, performing smoothing treatment on the inflammatory marker values to generate continuous inflammatory response trend lines, performing metabolic dynamics analysis on the serum drug concentration monitoring values to extract distribution volume and elimination rate constants, and generating individual metabolic feature parameter sets; The bacterial load power fitting module is used for performing differential operation on the continuous inflammatory response trend line to obtain an instant change slope and quantifying the instant change slope into bacterial activity, mapping the bacterial activity to a nonlinear growth characteristic curve to calculate the proliferation probability value of the bacterial population under the current loading capacity, and generating a bacterial load replication rate vector comprising the instant proliferation speed and proliferation saturation; The medicine fungus competition analysis module is used for calling individual metabolism characteristic parameter sets to be substituted into a concentration-time curve equation to calculate the medicine concentration of a target part at the current moment, converting the medicine concentration of the target part into a theoretical bacteriostasis rate through a concentration-effect mapping relation, and performing vector difference operation on the theoretical bacteriostasis rate and the instantaneous proliferation rate in a biomass replication rate vector to generate a net inhibition gradient value; The window-shifting-rate optimizing module is used for projecting the net inhibition gradient value to a drug-resistant risk interval defined by the minimum inhibition concentration and the mutation-preventing concentration, calculating the crossing detention time of the drug concentration curve falling in the drug-resistant risk interval through time integral operation, and executing the parameter reverse iterative optimizing with the minimized crossing detention time as a constraint condition to generate a drug administration parameter adjusting instruction when the crossing detention time exceeds a preset threshold.
  2. 2. The system for monitoring and managing data of an antibacterial drug administration process according to claim 1, wherein the specific process of collecting the serum drug concentration monitoring value and the inflammatory marker value in time series and performing smoothing on the inflammatory marker value to generate a continuous inflammatory response trend line is as follows: Constructing a sampling time index matrix, calculating the time stamp difference value of adjacent data points to generate an adaptive attenuation weight, substituting the adaptive attenuation weight into a weighted moving average algorithm to execute denoising operation on the inflammatory marker value, and removing outlier noise points with deviation exceeding a preset standard deviation multiple; And calling a cubic spline interpolation function to execute piecewise polynomial fitting on the denoised discrete data points, and generating a continuous inflammatory response trend line under the condition of keeping continuous constraint of the first derivative and the second derivative.
  3. 3. The system for monitoring and managing data of an antibacterial drug administration process according to claim 1, wherein the specific process for generating the individual metabolic characteristic parameter set by performing the metabolic analysis on the serum drug concentration monitoring value to extract the distribution volume and the elimination rate constant is as follows: Performing natural logarithmic transformation on the serum drug concentration monitoring value, identifying and intercepting a drug concentration monotonically decreasing interval as a linear regression interval, performing linear fitting on linear regression interval data points by adopting a least square method, and extracting the absolute value of the slope of a fitting straight line as an elimination rate constant; And reversely extending the fitting straight line to the administration time to obtain a theoretical initial concentration intercept, calculating the ratio of the administration dose to the theoretical initial concentration intercept to obtain an apparent distribution volume, and combining the elimination rate constant and the apparent distribution volume into an individual metabolism characteristic parameter set.
  4. 4. The system for monitoring and managing data of antibacterial drug administration process according to claim 1, wherein the specific process of obtaining the instant change slope and quantifying the instant change slope into the bacterial activity by performing differential operation on the continuous inflammatory response trend line is as follows: Setting a sliding differential window on a continuous inflammatory response trend line, and calculating a discrete difference quotient of a current time value and a previous time value in the window; Comparing the discrete difference quotient with a preset basic metabolic fluctuation threshold value, and screening the difference quotient exceeding the basic metabolic fluctuation threshold value as an effective proliferation signal; performing a maximum-minimum normalization process on the effective proliferation signal, mapping to a dimensionless interval of zero to one, generates bacterial activity.
  5. 5. The system for monitoring and managing data of an antibacterial drug administration process according to claim 4, wherein the specific process of mapping the bacterial activity to a nonlinear growth characteristic curve to calculate the proliferation probability value of the bacterial population under the current load, and generating a vector of the replication rate of the bacterial population including the instantaneous proliferation rate and the proliferation saturation is as follows: Taking a preset logic growth function as a nonlinear growth characteristic curve, inputting the bacterial activity as an independent variable into a first derivative equation of the logic growth function, and calculating a derivative value as an instantaneous proliferation speed; Calculating the numerical difference between the bacterial activity and the asymptote peak of the logistic growth function, differentiating the numerical difference into proliferation saturation, and constructing the instantaneous proliferation speed and the proliferation saturation into a two-dimensional array to generate a bacterial load replication rate vector.
  6. 6. The antibacterial drug administration process data monitoring and management system according to claim 1, wherein the specific process of substituting the individual metabolic characteristic parameter set into a concentration-time curve equation to calculate the target site drug concentration at the current moment and converting the target site drug concentration into the theoretical antibacterial rate through a concentration-effect mapping relationship is as follows: Extracting individual metabolic characteristic parameter set elimination rate constant and distribution volume, constructing a chamber model exponential decay equation by combining preset tissue penetration coefficients, and iteratively calculating interstitial fluid simulation concentration at the current moment as target site drug concentration; The S-shaped pharmacodynamic function containing the half maximum effect concentration and the Hill coefficient is taken as a concentration-effect mapping relation, and the concentration of the target part drug is input into the S-shaped pharmacodynamic function to calculate a pharmacodynamic intensity coefficient; And multiplying the drug effect intensity coefficient by a preset maximum bacteriostasis rate constant to generate a theoretical bacteriostasis rate.
  7. 7. The system for monitoring and managing data of an antibacterial drug administration process according to claim 6, wherein the specific process of generating a net inhibition gradient value by performing vector difference operation on the instantaneous proliferation rate in the vector of the theoretical antibacterial rate and the bacterial load replication rate is as follows: performing time sequence index alignment on the theoretical bacteriostasis rate and the biomass replication rate vector, and extracting a theoretical bacteriostasis rate value and an instantaneous proliferation rate at the same moment; And performing scalar subtraction operation, subtracting the instantaneous proliferation speed from the theoretical bacteriostasis rate value to obtain an instantaneous countermeasure difference value, and performing smoothing filtering processing on the continuous time sequence instantaneous countermeasure difference value by using an exponential weighted moving average algorithm to generate a net inhibition gradient value.
  8. 8. The system for monitoring and managing data of antibacterial drug administration process according to claim 1, wherein the specific process of projecting the net inhibition gradient value to a drug-resistant risk interval defined by the minimum inhibitory concentration and the mutation-preventing concentration and calculating the crossing residence time of the drug concentration curve in the drug-resistant risk interval by time integral operation is as follows: Constructing a numerical comparison logic, comparing the drug concentration of the target part with the minimum inhibitory concentration and the mutation-preventing concentration point by point, and marking discrete sampling points of the drug concentration of the target part between the minimum inhibitory concentration and the mutation-preventing concentration as intra-interval sampling points; Searching a net inhibition gradient value of a sampling point in a corresponding interval, and screening data points with the net inhibition gradient value smaller than a preset complete inhibition threshold value as effective retention sampling points; And performing accumulation summation on the time intervals of the effective retention sampling points by using a trapezoidal integration rule to generate the traversing retention time length.
  9. 9. The system for monitoring and managing data of an antibacterial drug administration process according to claim 8, wherein the specific process for generating the drug administration parameter adjustment instruction by performing the parameter reverse iterative optimization under the constraint of minimizing the traverse residence time when the traverse residence time exceeds a preset threshold value is as follows: Setting the administration dosage value and the administration interval time as variable optimization parameters, setting the minimized ride-through residence time as an objective function, and setting the peak concentration lower than the toxicity threshold as a boundary constraint condition; Starting a coordinate descent search algorithm, establishing an alternate iteration loop logic, in a single iteration step, keeping the dosing interval time constant, executing one-dimensional extremum search for the dosing quantity value to update the dosing quantity value, then keeping the updated dosing quantity value constant, executing one-dimensional extremum search for the dosing interval time to update the dosing interval time, and calculating the simulated traversing retention time under different parameter combinations; Stopping iteration when the simulated traversing residence time is converged to a preset convergence domain, and extracting the current optimal parameter combination code to be converted into the dosing parameter adjustment instruction.

Description

Antibacterial drug administration process data monitoring and management system Technical Field The invention relates to the technical field of medical information data processing, in particular to a system for monitoring and managing data of an antibacterial drug administration process. Background The continued evolution of bacterial resistance has become a serious challenge in the world public health field, severely affecting the therapeutic prognosis and patient safety of infectious diseases. In the clinical treatment process, the reasonable application of the antibacterial drugs is a key link for restraining the generation of drug-resistant strains and guaranteeing the medical quality. Along with the improvement of the medical informatization level, the medication process is finely managed and controlled by a digital means, and the balance of the drug treatment effect and the drug resistance risk control becomes the core development direction of a modern hospital management and clinical auxiliary decision-making system. However, the existing clinical medication monitoring method mainly relies on a static rule engine to audit medical advice, and has the substantial defects that only single-point verification can be carried out aiming at the dosage limit and the incompatibility at the moment of prescription making, and the real-time response of metabolic processes and pathogens after the medicine enters a human body can not be perceived dynamically. The prior art ignores the real-time influence of the physiological function fluctuation of the patient on the drug clearance rate, and lacks a quantitative analysis means for the dynamic competition relationship between the drug sterilization speed and the bacterial propagation speed. This results in the fact that in practice it often happens that the drug concentration, although numerically within the specification, does not cover the bacteria rapid proliferation phase or stay too long in the drug-resistant mutation window, thus causing the risk of failing the occult therapy and inducing screening of drug-resistant strains. Disclosure of Invention Aiming at the defects of the prior art, the invention provides a system for monitoring and managing the data of the antibacterial drug administration process, which solves the problems of the background technology. The invention is realized by the following technical scheme that the antibacterial drug administration process data monitoring and management system comprises a multidimensional feature mapping module, a bacterial load power fitting module, a drug bacteria competition analysis module and a drug bacteria competition analysis module, wherein the multidimensional feature mapping module is used for acquiring serum drug concentration monitoring values and inflammatory marker values according to time sequences, performing smoothing treatment on the inflammatory marker values to generate continuous inflammatory response trend lines, performing metabolic dynamics analysis on the serum drug concentration monitoring values to extract distribution volume and elimination rate constants to generate individual metabolism feature parameter sets, performing differential operation on the continuous inflammatory response trend lines to obtain instant change slope and quantifying the instant change slope to obtain bacterial activity, mapping the bacterial activity to proliferation probability values of a nonlinear growth feature curve under the current loading capacity to generate a bacterial load replication rate vector containing instant proliferation speed and proliferation saturation, the drug bacteria competition analysis module is used for substituting the target drug concentration at the current moment into a concentration-time curve equation, converting the target drug concentration into a theoretical antibacterial rate through the concentration-time curve mapping relation, performing differential operation on the instant proliferation rate vectors in the theoretical antibacterial rate and the bacterial activity curve to obtain instant change slope and quantifying the instant change probability value to obtain the bacterial activity value, mapping the bacterial activity value to map to calculate the bacterial activity value at the optimal value in the time-lag time interval by the optimal value from the optimal value to the anti-mutation risk curve, and executing the parameter reverse iterative optimization taking the minimized traversing residence time as a constraint condition to generate a dosing parameter adjustment instruction when the traversing residence time exceeds a preset threshold. The method comprises the steps of obtaining a serum drug concentration monitoring value and an inflammatory marker value according to a time sequence, performing smoothing processing on the inflammatory marker value to generate a continuous inflammatory response trend line, wherein the specific process comprises the steps of c