Intelligent infusion management method and system based on p-core needle

Abstract

The invention discloses an intelligent infusion management method and system based on a p-card needle, and relates to the technical field of intelligent infusion management, and the method comprises the steps: collecting physiological monitoring data of a patient, and carrying out the preprocessing; constructing a physiological monitoring voiceprint and a fitting model function relationship between the physiological monitoring voiceprint and the infusion speed; adjusting the infusion speed to a coupling point according to whether the coupling point exists in a fitting model corresponding to the actual condition in the set range of the current infusion speed in the current monitoring data and the matching degree of the physiological monitoring voiceprint; and calculating a physiological monitoring voiceprint error and an error change rate, processing the physiological monitoring voiceprint error and the error change rate through a fuzzy control algorithm and a genetic algorithm, and then outputting a control signal. The relation curve of the infusion speed and the physiological monitoring voiceprint is fitted through a polynomial function, the infusion speed is regulated and controlled through the PID controller according to the received signal, and the infusion speed is optimized and controlled again in a fuzzy control and PID control linkage mode.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent infusion management, and more particularly to an intelligent infusion management method and system based on a p-care needle. Background Art

[0002] Intravenous infusion is a common medical procedure, but traditional infusion methods present numerous challenges, including a lack of real-time monitoring. Traditional systems cannot monitor key parameters such as intravascular pressure and infusion rate, potentially leading to problems such as overly rapid or slow infusions or needle dislodgement. Manual operation carries significant risks: Medical staff must manually adjust the infusion rate, which is labor-intensive and prone to errors. Safety issues: Traditional needle designs pose a risk of puncture wounds and are unable to detect infusion anomalies (such as vascular blockage or air bubbles) in real time. Low intelligence: Existing infusion systems lack intelligent algorithms and are unable to dynamically adjust infusion plans based on the patient's physiological state. Traditional infusion procedures often lead to a series of problems due to varying vascular conditions and inappropriate infusion rates. For example, patients with thin or fragile blood vessels are susceptible to vascular damage with standard needles. Excessively rapid infusion rates can increase cardiac stress, while excessively slow infusion rates can compromise treatment efficiency. Similarly, while the P-CARE needle can reduce vascular damage, there is room for improvement in infusion pressure and rate control.

[0003] Therefore, how to propose an intelligent infusion management method and system based on the P-CARE needle, and achieve precise infusion pressure and speed control on the basis of using the P-CARE needle to reduce vascular damage, and improve the safety and effectiveness of infusion is an urgent problem that technicians in this field need to solve. Summary of the Invention

[0004] In view of this, the present invention provides an intelligent infusion management method and system based on the P-CARE needle. On the basis of using the P-CARE needle to reduce vascular damage, it realizes precise infusion pressure and speed control, and improves the safety and effectiveness of infusion. To achieve the above purpose, the present invention adopts the following technical solutions:

[0005] An intelligent infusion management method based on a p-care needle, comprising:

[0006] Collect the patient's physiological monitoring data, perform preprocessing, and obtain processed monitoring data;

[0007] Constructing physiological monitoring voiceprints based on the processed monitoring data;

[0008] Establish a fitting model function relationship between physiological monitoring voiceprint and infusion rate;

[0009] Obtain the current monitoring data, cluster it to get the actual situation, and adjust the infusion speed to the coupling point based on whether there is a coupling point in the fitting model corresponding to the actual situation within the set range of the current infusion speed in the current monitoring data and the matching degree of the physiological monitoring voiceprint;

[0010] The physiological monitoring voiceprint measured at the coupling point is compared with the set physiological monitoring voiceprint to obtain the physiological monitoring voiceprint error, and the error change rate is calculated. The physiological monitoring voiceprint error and error change rate are processed by fuzzy control algorithm and genetic algorithm, and then the control signal is output to optimize the control of infusion speed.

[0011] Optionally, the patient's physiological monitoring data includes: intravascular pressure, heart rate, blood pressure, pressure fluctuation characteristics, infusion flow, nearby tissue pressure changes and local temperature data.

[0012] Optionally, the preprocessing includes: eliminating false values, including eliminating illegal values or outliers; eliminating abnormal working values, including eliminating data with negative values, and normalizing the data after elimination.

[0013] Optionally, after obtaining the processed monitoring data, the method further includes dividing the processed monitoring data into multiple groups of data corresponding to different situations using a clustering algorithm, and constructing a physiological monitoring voiceprint based on the data corresponding to different situations.

[0014] Optionally, the physiological monitoring voiceprint includes:

[0015] The combined signal formed by the intravascular pressure signal, heart rate signal and blood pressure signal indicates the strength of the tolerance under the time series. The intensity of the corresponding signal during infusion under different time series is collected. Assuming that the measured time point L i By sensor Q j Measured point L i The joint signal strength is expressed as:

[0016] {S ij (t), t=1,…,k(k≥1)};

[0017] Among them, S ij (t) represents the signal strength of the t-th measurement, k represents the number of measurements, i represents the i-th measured time point, j represents the j-th sensor, and the signal strength of the i-th measured time point is obtained by the j-th sensor to construct the joint signal fingerprint library F:

[0018]

[0019] Where m is the number of sensors, n is the number of signal groups, and the i-th row vector F i =[S i1 ,Si2 ,...,S im ] is the m sensors’ measured data about the measured point L i Fingerprint feature values are measured by different types of sensors to form multi-dimensional feature dimensions. i The tolerance when performing infusion at a certain point is the strength of the joint signal constructed by the information measured by each sensor.

[0020] Optionally, establishing a fitting model functional relationship between the physiological monitoring voiceprint and the infusion rate includes: based on a polynomial function and introducing an exponential forgetting factor, establishing a functional relationship between the physiological monitoring voiceprint and the infusion rate in each set of data to obtain multiple fitting models.

[0021] Optionally, adjusting the infusion speed to the coupling point includes: obtaining current monitoring data, clustering to obtain the actual situation, and obtaining the minimum point closest to the current infusion speed within the set range of the current infusion speed in the current monitoring data based on the fitting model corresponding to the actual situation. If the physiological monitoring voiceprint of the minimum point is smaller than the current physiological monitoring voiceprint obtained according to the current infusion speed, then adjusting the current infusion speed to the minimum point; otherwise, keeping the current infusion speed unchanged.

[0022] Optionally, the objective function for optimizing the control of the infusion rate is:

[0023] The objective function is

[0024] Where, e(t) is the system error, r(t) is the output of the PID controller, and t y is the rise time, o1, o2, and o3 are weights.

[0025] Optionally, the genetic algorithm includes: when using the genetic algorithm to find the optimal solution, adopting the following method: assuming that B1 is the optimal individual of the parent generation and B2 is the optimal individual of the offspring, when B1 is better than B2, retain B1; otherwise retain B2.

[0026] Optionally, an intelligent infusion management system based on the p-care needle includes:

[0027] Acquisition module: used to collect the patient's physiological monitoring data, perform preprocessing, and obtain processed monitoring data;

[0028] Physiological monitoring voiceprint construction module: used to construct physiological monitoring voiceprint based on the processed monitoring data;

[0029] Fitting module: used to establish the fitting model function relationship between physiological monitoring voiceprint and infusion speed;

[0030] Pre-adjustment module: used to obtain the current monitoring data, cluster the actual situation, and adjust the infusion speed to the coupling point according to whether there is a coupling point in the fitting model corresponding to the actual situation within the set range of the current infusion speed in the current monitoring data and the matching degree of the physiological monitoring voiceprint;

[0031] Feedback optimization and adjustment module: used to compare the physiological monitoring voiceprint measured at the coupling point with the set physiological monitoring voiceprint, obtain the physiological monitoring voiceprint error, and calculate the error change rate. After processing the physiological monitoring voiceprint error and error change rate through fuzzy control algorithm and genetic algorithm, the control signal is output to optimize the control of the infusion speed.

[0032] It can be seen from the above technical solutions that, compared with the prior art, the present invention provides an intelligent infusion management method and system based on the p-care needle, which has the following beneficial effects:

[0033] The present invention proposes an intelligent infusion management method based on a p-care needle, comprising: collecting physiological monitoring data of the patient, performing preprocessing to obtain processed monitoring data; constructing a physiological monitoring voiceprint based on the processed monitoring data; establishing a fitting model function relationship between the physiological monitoring voiceprint and the infusion speed; obtaining current monitoring data, clustering to obtain the actual situation, and adjusting the infusion speed to the coupling point according to whether there is a coupling point in the fitting model corresponding to the actual situation within the set range of the current infusion speed in the current monitoring data and the matching degree of the physiological monitoring voiceprint; comparing the physiological monitoring voiceprint measured at the coupling point with the set physiological monitoring voiceprint to obtain the physiological monitoring voiceprint error, and calculating the error change rate, processing the physiological monitoring voiceprint error and the error change rate through a fuzzy control algorithm and a genetic algorithm, and then outputting a control signal to optimize the control of the infusion speed. The present invention uses a polynomial function to fit the relationship curve between infusion rate and physiological monitoring soundprint. At fixed time intervals, a new model is fitted based on the new data collected, so that the model is iteratively updated in real time and provides accurate infusion rate fine-tuning suggestions. On this basis, the infusion rate is regulated by a PID controller based on the received signal. The infusion rate is controlled by fuzzy control and PID control in a linked manner, and the three parameters of the PID controller are dynamically generated online. The physiological monitoring soundprint is obtained from the sensor and compared with the set physiological monitoring soundprint to obtain the physiological monitoring soundprint error. The error change rate is also calculated. The physiological monitoring soundprint and the error change rate are then quantified and fuzzified into fuzzy quantities. The fuzzy quantities are then fuzzy-determined based on the fuzzy inference principle to obtain the fuzzy control quantity as the controlled object. The genetic algorithm is then used to select a suitable fitness function in the optimization search to search the fitness values of the three parameters of the PID controller, thereby increasing the convergence speed of the genetic algorithm and calculating the optimal solution, thereby achieving optimal infusion rate control. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0035] Figure 1 This is a schematic flow chart of an intelligent infusion management method based on a p-care needle provided by the present invention.

[0036] Figure 2 This is a structural framework diagram of an intelligent infusion management system based on the p-care needle provided by the present invention. DETAILED DESCRIPTION

[0037] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0038] The embodiment of the present invention discloses an intelligent infusion management method based on p-care needles, such as Figure 1 As shown, including:

[0039] Collect the patient's physiological monitoring data, perform preprocessing, and obtain processed monitoring data;

[0040] Constructing physiological monitoring voiceprints based on the processed monitoring data;

[0041] Establish a fitting model function relationship between physiological monitoring voiceprint and infusion rate;

[0042] Obtain the current monitoring data, cluster it to get the actual situation, and adjust the infusion speed to the coupling point based on whether there is a coupling point in the fitting model corresponding to the actual situation within the set range of the current infusion speed in the current monitoring data and the matching degree of the physiological monitoring voiceprint;

[0043] The physiological monitoring voiceprint measured at the coupling point is compared with the set physiological monitoring voiceprint to obtain the physiological monitoring voiceprint error, and the error change rate is calculated. The physiological monitoring voiceprint error and error change rate are processed by fuzzy control algorithm and genetic algorithm, and then the control signal is output to optimize the control of infusion speed.

[0044] Furthermore, the patient's physiological monitoring data includes: intravascular pressure, heart rate, blood pressure, pressure fluctuation characteristics, infusion flow, nearby tissue pressure changes and local temperature data.

[0045] Furthermore, the preprocessing includes: eliminating false values, including eliminating illegal values or outliers; eliminating abnormal working values, including eliminating data with negative values, and normalizing the data after elimination.

[0046] Furthermore, after obtaining the processed monitoring data, the method further includes dividing the processed monitoring data into multiple groups of data corresponding to different situations using a clustering algorithm, and constructing a physiological monitoring voiceprint based on the data corresponding to different situations.

[0047] In a specific embodiment, the use of a clustering algorithm to divide the processed monitoring data into multiple groups of data corresponding to different situations includes:

[0048] For example, vascular pressure is divided into six segments. By monitoring the average heart rate and blood pressure signals at these six levels, a total of six different scenarios can be generated. Using the K-means clustering algorithm, the two-dimensional data consisting of vascular pressure, heart rate, and blood pressure signals is segmented to generate scenarios 1 through 6.

[0049] Furthermore, the physiological monitoring voiceprint includes:

[0050] The combined signal formed by the intravascular pressure signal, heart rate signal and blood pressure signal indicates the strength of the tolerance under the time series. The intensity of the corresponding signal during infusion under different time series is collected. Assuming that the measured time point L i By sensor Q j Measured point L i The joint signal strength is expressed as:

[0051] {S ij (t), t=1,…,k(k≥1)};

[0052] Among them, S ij (t) represents the signal strength of the t-th measurement, k represents the number of measurements, i represents the i-th measured time point, j represents the j-th sensor, and the signal strength of the i-th measured time point is obtained by the j-th sensor to construct the joint signal fingerprint library F:

[0053]

[0054] Where m is the number of sensors, n is the number of signal groups, and the i-th row vector F i =[S i1 ,S i2 ,...,Sim ] is the m sensors’ measured data about the measured point L i Fingerprint feature values are measured by different types of sensors to form multi-dimensional feature dimensions. i The tolerance when performing infusion at a certain point is the strength of the joint signal constructed by the information measured by each sensor.

[0055] Furthermore, establishing a fitting model functional relationship between the physiological monitoring voiceprint and the infusion rate includes: based on a polynomial function and introducing an exponential forgetting factor, establishing a functional relationship between the physiological monitoring voiceprint and the infusion rate in each set of data to obtain multiple fitting models.

[0056] In a specific embodiment, the polynomial function is used to calculate the physiological monitoring voiceprint f(x) based on the infusion rate x. Let the actual physiological monitoring voiceprint be f. When the error between the polynomial function f(x) and f is minimized, the fitting effect is optimal. The polynomial function that minimizes the sum of squared errors is the best square approximation polynomial function of f. Let M be the polynomial order, the set of all M-order polynomial functions be S, where any polynomial function is s, and the best square approximation polynomial function is s*, then:

[0057] Here, ||*||2 represents the 2-norm.

[0058] Suppose a set of basis functions of the polynomial function is β0(x),β1(x),β2(x),...,β M (x), then any M-order polynomial function s can be expressed as:

[0059] Among them, a represents the coefficient of the polynomial, a i represents the i-th basis function β i (x)The coefficient in front.

[0060] The exponential forgetting factor is related to time and is calculated as follows:

[0061]

[0062] Among them, x j represents the infusion rate in the jth processed monitoring data, Δt j Represents x j The absolute value of the time difference between the acquisition time and the current time, κ represents the preset attenuation factor, 0.01≤κ≤0.1;

[0063] By establishing the inner product matrix and the constant term matrix of the polynomial function, solving the coefficient matrix of the fitting model;

[0064] Based on the exponential forgetting factor and the value of the inner product matrix (β i (x),β k (x)) N , recursively obtain the value in the inner product matrix corresponding to the N+1th infusion rate (β i (x),β k (x)) N+1 , the calculation formula is:

[0065]

[0066] Among them, 0≤i≤M,0≤k≤M, M represents the order of the polynomial, β i (x) and β k (x) represents the basis function of the polynomial function, x N+1 Indicates the N+1th infusion rate, (t N+1 -t N ) represents the acquisition time difference between the N+1th infusion rate and the Nth infusion rate, (β i (x),β k (x)) represents the value of the i+1th row and k+1th column in the inner product matrix;

[0067] Based on the exponential forgetting factor and the value (θ i (x),β k (x)) N Recursively obtain the value of the constant term matrix corresponding to the N+1th physiological monitoring voiceprint (θ i (x),β k (x)) N+1 , the calculation formula is:

[0068]

[0069] Among them, θ(x) represents the physiological monitoring voiceprint, θ(x N+1 ) represents the N+1th physiological monitoring voiceprint, (θ i (x),β k (x)) represents the value of the k+1th row in the constant term matrix.

[0070] Basis function β k (x) Selecting Legendre polynomials as basis functions can improve the condition number of the inner product matrix and make the fitting results more accurate. Therefore, Legendre polynomials are used as basis functions and are defined as follows:

[0071]

[0072] Based on the above formula, the inner product matrix and the constant term matrix are calculated, the coefficients in the coefficient matrix are obtained, and the fitting model is obtained.

[0073] Furthermore, the adjustment of the infusion speed to the coupling point includes: obtaining current monitoring data, clustering to obtain the actual situation, and according to the fitting model corresponding to the actual situation, obtaining the minimum point closest to the current infusion speed within the set range of the current infusion speed in the current monitoring data; if the physiological monitoring voiceprint of the minimum point is smaller than the current physiological monitoring voiceprint obtained according to the current infusion speed, then adjusting the current infusion speed to the minimum point; otherwise, keeping the current infusion speed unchanged.

[0074] Specifically, the coupling point includes a minimum point of a function obtained by derivation, and an infusion speed that minimizes the physiological monitoring voiceprint is obtained.

[0075] Furthermore, the objective function for optimizing the control of the infusion rate is:

[0076] The objective function is

[0077] Where, e(t) is the system error, r(t) is the output of the PID controller, and t y is the rise time, o1, o2, and o3 are weights.

[0078] In a specific embodiment, the physiological monitoring voiceprint measured at the coupling point is compared with the set physiological monitoring voiceprint to obtain the physiological monitoring voiceprint error, and the error change rate is calculated. The physiological monitoring voiceprint error and the error change rate are processed by a fuzzy control algorithm and a genetic algorithm to output a control signal. The optimized control of the infusion rate specifically includes:

[0079] The parameter selections are as follows:

[0080] (1) Three basic parameters of PID controller:

[0081] The three basic parameters of the PID controller are set to values in the range of Vp1 = [0, 10], Vi1 = [0, 1], and Vd1 = [0, 2];

[0082] (2) Physiological monitoring voiceprint error E and error change rate Ec:

[0083] The domain levels of physiological monitoring voiceprint error E and error change rate Ec are: E, Ec = {-6, -5, -4, -3, -2, -1, 0, 1, 2, 3, 4, 5, 6};

[0084] (3) Three parameters of fuzzy control:

[0085] The domain of the three parameters of fuzzy control is: Vp, Vi, Vd = {-6, -5, -4, -3, -2, -1, 0, 1, 2, 3, 4, 5, 6};

[0086] (4) Parameters after genetic algorithm optimization:

[0087] The three parameters after genetic algorithm optimization are: Vp = Vp1 + △V1*V1, Vi = Vi2 + △V2*V2, Vd = Vd3 + △V3*V3, among which Vp1, Vi1, and Vd1 are the preset values of the three parameters of the PID controller, △V1, △V2, and △V3 are the outputs after fuzzy control, and V1, V2, and V3 are the three proportional parameters calculated by the genetic algorithm;

[0088] (5) Proportional parameters:

[0089] The value ranges of the three scale parameters are V1 = [0.6, 6], V2 = [0.6, 2], and V3 = [0.6, 2];

[0090] (6) Objective function:

[0091] The objective function is

[0092] Where, e(t) is the system error, r(t) is the output of the PID controller, and t y is the rise time, o1, o2, o3 are weights;

[0093] (7) Fitness function:

[0094] The fitness function is the inverse of the objective function, and the fitness function U=1 / D.

[0095] Furthermore, the genetic algorithm includes: when using the genetic algorithm to find the optimal solution, the following method is adopted: assuming that B1 is the optimal individual of the parent generation and B2 is the optimal individual of the offspring generation, when B1 is better than B2, retain B1; otherwise retain B2.

[0096] Furthermore, PWM pulse width modulation is used to adjust the infusion speed by adjusting the PWM duty cycle. When the infusion speed is lower than the expected value, the duty cycle is increased; otherwise, the duty cycle is reduced.

[0097] Furthermore, an intelligent infusion management system based on p-care needles, such as Figure 2 As shown, including:

[0098] Acquisition module: used to collect the patient's physiological monitoring data, perform preprocessing, and obtain processed monitoring data;

[0099] Physiological monitoring voiceprint construction module: used to construct physiological monitoring voiceprint based on the processed monitoring data;

[0100] Fitting module: used to establish the fitting model function relationship between physiological monitoring voiceprint and infusion speed;

[0101] Pre-adjustment module: used to obtain the current monitoring data, cluster the actual situation, and adjust the infusion speed to the coupling point according to whether there is a coupling point in the fitting model corresponding to the actual situation within the set range of the current infusion speed in the current monitoring data and the matching degree of the physiological monitoring voiceprint;

[0102] Feedback optimization and adjustment module: used to compare the physiological monitoring voiceprint measured at the coupling point with the set physiological monitoring voiceprint, obtain the physiological monitoring voiceprint error, and calculate the error change rate. After processing the physiological monitoring voiceprint error and error change rate through fuzzy control algorithm and genetic algorithm, the control signal is output to optimize the control of the infusion speed.

[0103] In a specific embodiment, an intelligent infusion management device based on the p-CARE needle includes: An intelligent pressure sensing module: A micro-pressure sensor is integrated into the proximal end of the P-CARE needle to monitor the pressure exerted on the needle in the blood vessel in real time. If the pressure exceeds a preset safety range, whether due to blood vessel blockage, needle displacement, or excessive infusion speed, the sensor quickly detects the pressure and transmits the signal to the control unit.

[0104] Infusion speed control device: This device consists of a motor, gear train, and infusion tube clamp. The control unit, based on signals from the pressure sensing module, controls the motor and, through the gear train, adjusts the tightness of the infusion tube clamp to precisely adjust the infusion speed. For example, if pressure rises, indicating an excessively fast infusion rate, the motor drives the gear train to tighten the tube clamp, slowing the infusion rate; otherwise, the clamp is loosened.

[0105] Data Processing and Control Unit: This unit utilizes a high-performance microprocessor to receive signals from the pressure sensing module, analyze and process them based on built-in algorithms, and send instructions to the infusion rate control device and display module. The unit also stores infusion data, such as pressure change curves and infusion rate adjustment records, for easy review by medical staff.

[0106] Display and Alarm Module: Equipped with an LCD screen, it displays key information such as infusion rate, pressure value, and remaining infusion volume in real time. When pressure is abnormal or other faults occur, the alarm module immediately issues an audible and visual alarm to alert medical staff to take timely action.

[0107] Specifically, the operating steps of the intelligent infusion management device based on the P-CARE needle include: Assembly and preparation: Connecting and securing the intelligent pressure sensing module to the P-CARE needle to ensure that the pressure sensor can accurately sense the pressure at the needle tip; Installing the infusion rate control device at a suitable location on the infusion tube near the needle tip; Connecting the control unit and the display and alarm module; Performing a power-on test; and Calibrating the pressure and rate parameters.

[0108] Infusion process: The P-CARE needle is inserted into the patient's blood vessel as usual, and the infusion system is activated. The intelligent pressure sensing module monitors pressure in real time and transmits the data to the control unit. The control unit analyzes and processes the data based on a preset algorithm. If the pressure is normal, the infusion rate is maintained at the set value. If the pressure is abnormal, the control unit sends a command to the infusion rate control device to adjust the infusion rate and displays an alarm on the display and alarm module.

[0109] Maintenance and Calibration: Regularly calibrate the pressure sensing module to ensure accurate pressure monitoring. Check the operating conditions of the mechanical components of the infusion speed control device, such as the motor and gear set, and clean and maintain them promptly to ensure stable operation of the infusion system.

[0110] In a specific embodiment, the system also includes: combining dynamic algorithms (sliding window analysis, machine learning) with the patient's physiological monitoring data: intravascular pressure, heart rate, blood pressure, pressure fluctuation characteristics, infusion flow rate, nearby tissue pressure changes and local temperature data to determine abnormalities, such as needle dislodgement, blood vessel blockage, and air bubble ingress. Needle dislodgement: Rapidly identified by a sudden drop in intravascular pressure (near zero); blood vessel blockage: A comprehensive judgment based on a continuous increase in pressure and a decrease in flow rate; Air bubble ingress: Dual verification through pressure fluctuation characteristics and flow rate changes; Drug extravasation: Early warning through tissue pressure changes and local temperature monitoring. Active intervention (such as automatic speed reduction, infusion cessation, or bubble removal) is performed when abnormalities are detected.

[0111] The present invention uses a polynomial function to fit the relationship curve between infusion rate and physiological monitoring soundprint. At fixed time intervals, a new model is fitted based on the new data collected, so that the model is iteratively updated in real time and provides accurate infusion rate fine-tuning suggestions. On this basis, the infusion rate is regulated by a PID controller based on the received signal. The infusion rate is controlled by fuzzy control and PID control in a linked manner, and the three parameters of the PID controller are dynamically generated online. The physiological monitoring soundprint is obtained from the sensor and compared with the set physiological monitoring soundprint to obtain the physiological monitoring soundprint error. The error change rate is also calculated. The physiological monitoring soundprint and the error change rate are then quantified and fuzzified into fuzzy quantities. The fuzzy quantities are then fuzzy-determined based on the fuzzy inference principle to obtain the fuzzy control quantity as the controlled object. The genetic algorithm is then used to select a suitable fitness function in the optimization search to search the fitness values of the three parameters of the PID controller, thereby increasing the convergence speed of the genetic algorithm and calculating the optimal solution, thereby achieving optimal infusion rate control.

[0112] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. For the devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.

[0113] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An intelligent infusion management method based on p-care needle, characterized in that: include: Collect the patient's physiological monitoring data, perform preprocessing, and obtain processed monitoring data; Constructing physiological monitoring voiceprints based on the processed monitoring data; Establish a fitting model function relationship between physiological monitoring voiceprint and infusion rate; Obtain the current monitoring data, cluster it to get the actual situation, and adjust the infusion speed to the coupling point based on whether there is a coupling point in the fitting model corresponding to the actual situation within the set range of the current infusion speed in the current monitoring data and the matching degree of the physiological monitoring voiceprint; The physiological monitoring voiceprint measured at the coupling point is compared with the set physiological monitoring voiceprint to obtain the physiological monitoring voiceprint error, and the error change rate is calculated. The physiological monitoring voiceprint error and error change rate are processed by fuzzy control algorithm and genetic algorithm, and then the control signal is output to optimize the control of infusion speed.

2. The intelligent infusion management method based on the p-care needle according to claim 1, characterized in that: The patient's physiological monitoring data includes: intravascular pressure, heart rate, blood pressure, pressure fluctuation characteristics, infusion flow, nearby tissue pressure changes and local temperature data.

3. The intelligent infusion management method based on p-care needle according to claim 1, characterized in that: The preprocessing includes: eliminating false values, including eliminating illegal values or outliers; eliminating abnormal working values, including eliminating data with negative values, and normalizing the data after elimination.

4. The intelligent infusion management method based on p-care needle according to claim 1, characterized in that: After obtaining the processed monitoring data, the method further includes dividing the processed monitoring data into multiple groups of data corresponding to different situations using a clustering algorithm, and constructing a physiological monitoring voiceprint based on the data corresponding to different situations.

5. The intelligent infusion management method based on p-care needle according to claim 1, characterized in that: The physiological monitoring voiceprint includes: The combined signal formed by the intravascular pressure signal, heart rate signal and blood pressure signal indicates the strength of the tolerance under the time series. The intensity of the corresponding signal during infusion under different time series is collected. Assuming that the measured time point L i By sensor Q j Measured point L i The joint signal strength is expressed as: {S ij (t),t=1,…,k(k≥1)}; Among them, S ij (t) represents the signal strength of the t-th measurement, k represents the number of measurements, i represents the i-th measured time point, j represents the j-th sensor, and the signal strength of the i-th measured time point is obtained by the j-th sensor to construct the joint signal fingerprint library F: Where m is the number of sensors, n is the number of signal groups, and the i-th row vector F i =[S i1 ,S i2 ,...,S im ] is the m sensors’ measured data about the measured point L i Fingerprint feature values are measured by different types of sensors to form multi-dimensional feature dimensions. i The tolerance when performing infusion at a certain point is the strength of the joint signal constructed by the information measured by each sensor.

6. The intelligent infusion management method based on p-care needle according to claim 1, characterized in that: The establishment of the fitting model functional relationship between the physiological monitoring voiceprint and the infusion speed includes: based on a polynomial function and introducing an exponential forgetting factor, establishing a functional relationship between the physiological monitoring voiceprint and the infusion speed in each set of data to obtain multiple fitting models.

7. The intelligent infusion management method based on p-care needle according to claim 1, characterized in that: The adjusting of the infusion speed to the coupling point includes: obtaining current monitoring data, clustering to obtain the actual situation, and obtaining the minimum point closest to the current infusion speed within the set range of the current infusion speed in the current monitoring data based on the fitting model corresponding to the actual situation. If the physiological monitoring voiceprint of the minimum point is smaller than the current physiological monitoring voiceprint obtained according to the current infusion speed, then adjusting the current infusion speed to the minimum point; otherwise, keeping the current infusion speed unchanged.

8. The intelligent infusion management method based on p-care needle according to claim 1, characterized in that: The objective function of optimizing the infusion speed is: The objective function is Where, e(t) is the system error, r(t) is the output of the PID controller, and t y is the rise time, o1, o2, and o3 are weights.

9. The intelligent infusion management method based on p-care needle according to claim 1, characterized in that: The genetic algorithm includes: when using the genetic algorithm to find the optimal solution, the following method is adopted: assuming that B1 is the optimal individual of the parent generation and B2 is the optimal individual of the offspring generation, when B1 is better than B2, retain B1; otherwise retain B2.

10. An intelligent infusion management system based on p-care needles, characterized in that: include: Acquisition module: used to collect the patient's physiological monitoring data, perform preprocessing, and obtain processed monitoring data; Physiological monitoring voiceprint construction module: used to construct physiological monitoring voiceprint based on the processed monitoring data; Fitting module: used to establish the fitting model function relationship between physiological monitoring voiceprint and infusion speed; Pre-adjustment module: used to obtain the current monitoring data, cluster the actual situation, and adjust the infusion speed to the coupling point according to whether there is a coupling point in the fitting model corresponding to the actual situation within the set range of the current infusion speed in the current monitoring data and the matching degree of the physiological monitoring voiceprint; Feedback optimization and adjustment module: used to compare the physiological monitoring voiceprint measured at the coupling point with the set physiological monitoring voiceprint, obtain the physiological monitoring voiceprint error, and calculate the error change rate. After processing the physiological monitoring voiceprint error and error change rate through fuzzy control algorithm and genetic algorithm, the control signal is output to optimize the control of the infusion speed.

Images