Device for monitoring blood leakage by using surface acoustic waves in blood purification system

Abstract

The invention discloses a device for realizing blood leakage monitoring by using surface acoustic waves in a blood purification system, which comprises a sound wave information detection module, an early warning model design module, a sensing performance evaluation module and a remote response optimization module, and is characterized in that a human body blood vessel distribution model is constructed and a blood leakage monitoring early warning model is designed; the blood leakage prediction risk degree is evaluated, human blood vessel blood leakage risk marking is carried out, blood leakage risk monitoring of different puncture points of hemodialysis is achieved, pertinence and accuracy of blood leakage monitoring are guaranteed, and then the blood leakage risk of different puncture points of hemodialysis is monitored by comprehensively evaluating the sensing performance of the surface acoustic wave filter and introducing interference factor analysis. Therefore, risk prediction and device optimization management are carried out on the sensing performance of the surface acoustic wave filter in a long-term use state, continuous and stable sensing performance of the surface acoustic wave filter is ensured, and finally blood leakage monitoring with high pertinence, high accuracy and good stability in a blood purification system is realized by using surface acoustic waves.

Description

Technical Field

[0001] The invention relates to the technical field of acoustic wave monitoring, and in particular to a device for realizing blood leakage monitoring by using surface acoustic waves in a blood purification system. Background Art

[0002] Surface acoustic wave filter (SAWF) is a device that works on the principle of piezoelectric effect. Its basic structure is to make two groups of acoustic-electric transducers and two groups of interdigital transducers on the polished surface of the substrate material with piezoelectric characteristics, which are used as transmitting transducers and receiving transducers respectively. Among them, the two groups of interdigital transducers transmit and receive acoustic wave signals respectively, taking the transmitted acoustic wave signal as the input signal and the received acoustic wave signal as the output signal to realize the transmission of acoustic wave signals. The two groups of acoustic-electric transducers perform electro-acoustic-acoustic-electric conversion respectively. The process of detecting waste liquid of a blood purification system by using a surface acoustic wave filter is as follows: injecting a blood detection reagent, for example, a mixed reagent of ellagic acid and calcium chloride, into the waste liquid. When the waste liquid contains blood, the blood will produce a coagulation reaction with the detection reagent, thereby producing a solid substance, so that the sound wave signal output by the surface acoustic wave filter during the waste liquid detection will change; The existing technology of using surface acoustic waves to monitor blood leakage in blood purification systems has the defects of insufficient pertinence and stability of blood leakage monitoring. Due to the differences in the puncture points of hemodialysis, the blood flow rate and thrombosis risk of puncture points in different parts are different, which will lead to different predicted probabilities of blood leakage risks at the puncture points, resulting in insufficient pertinence of blood leakage monitoring and difficulty in predicting the pertinence risk of specific puncture points. Moreover, when the filter is in long-term use, electromagnetic interference, mechanical vibration and ambient temperature will cause the filter sensing performance to decline, resulting in attenuation and delay of acoustic wave signal monitoring, resulting in insufficient stability of the filter sensing performance for blood leakage monitoring. In view of the above technical defects, a solution is now proposed. Summary of the invention

[0003] The purpose of the present invention is to solve the defects of insufficient pertinence and stability of blood leakage monitoring in the prior art, design a blood leakage monitoring and early warning model by constructing a human blood vessel distribution model and detecting the acoustic wave information of waste liquid in characteristic areas of the human body, evaluate the risk degree of blood leakage prediction, and perform corresponding human blood vessel blood leakage risk marking, thereby realizing blood leakage risk monitoring of different puncture points of hemodialysis and ensuring the pertinence and accuracy of blood leakage monitoring; by comprehensively evaluating the sensing performance of the surface acoustic wave filter, and then introducing interference factors to analyze the influence of the filter sensing performance, the sensing performance of the surface acoustic wave filter is predicted under long-term use, the aging degree of the surface acoustic wave filter is evaluated and the device is optimized and managed, so as to ensure that the sensing performance of the surface acoustic wave filter is continuously stable, and finally realize the blood leakage monitoring with strong pertinence, high accuracy and good stability using surface acoustic waves in the blood purification system.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions: A device for realizing blood leakage monitoring using surface acoustic waves in a blood purification system, comprising an acoustic wave information detection module, an early warning model design module, a sensor performance evaluation module and a remote response optimization module, wherein the acoustic wave information detection module, the early warning model design module, the sensor performance evaluation module and the remote response optimization module are communicatively connected; The acoustic wave information detection module is used to detect the acoustic wave information of the collected waste liquid: construct a human blood vessel distribution model and mark N0 characteristic areas, detect the waste liquid in the blood purification process through the surface acoustic wave filter, detect the waste liquid in N0 characteristic areas in real time, and collect the corresponding acoustic wave information; The early warning model design module is used to design a blood leakage monitoring early warning model: by analyzing and processing the acoustic wave information of N0 characteristic areas, a blood leakage monitoring early warning model is designed, the degree of blood leakage prediction risk in the characteristic areas is evaluated, and corresponding risk warning signals are generated, and then risk marking is performed on the characteristic areas of the human blood vessel distribution model; The sensing performance evaluation module is used to evaluate the sensing performance of the surface acoustic wave filter: by analyzing the timing changes of the acoustic wave signal, the stability of the acoustic wave signal is evaluated, and by monitoring the response delay of the acoustic wave signal, the real-time performance of the acoustic wave signal is evaluated. Then, the sensing performance of the surface acoustic wave filter is comprehensively evaluated through the stability and real-time performance of the acoustic wave signal. The remote response optimization module is used for remote blood leakage monitoring and device optimization management: by monitoring and analyzing the impact of interference factors on the electronic components of the surface acoustic wave filter, the risk of the sensing performance of the surface acoustic wave filter is predicted, the aging degree of the surface acoustic wave filter device is evaluated and corresponding early warning prompts are issued, thereby optimizing the device management.

[0005] Furthermore, the specific design process of the blood leakage monitoring and early warning model is as follows: S2-1, construct a comparison check sub-model to pre-process the sound wave information: The surface acoustic wave filter for real-time detection of waste liquid is marked as an experimental sensor P, and a redundant surface acoustic wave filter is set to perform regular verification of the waste liquid, the regular verification period is marked as Tj, and the redundant surface acoustic wave filter is marked as a verification sensor Q. By comparing the experimental sensor P with the verification sensor Q, the data of the experimental sensor P is verified; The sound wave information includes the frequency and amplitude of the sound wave signal, and the sound wave signal includes an input signal and an output signal; The input signals of the experimental sensor P and the verification sensor Q are set to be the same. After the experimental sensor P and the verification sensor Q monitor the waste liquid, the corresponding output signals are obtained respectively. By comparing the acoustic wave information of the two sets of output signals at the same time node, the frequency deviation and amplitude deviation between the experimental sensor P and the verification sensor Q are evaluated, thereby evaluating the accuracy of the acoustic wave signal of the experimental sensor P, determining the abnormality of the acoustic wave information, and performing corresponding verification processing; S2-2, then conduct a comprehensive analysis on the acoustic wave information of the pre-processed experimental sensor P: Mark any characteristic region as e, mark the surface acoustic wave filter of the characteristic region e as the experimental sensor Pe, and compare the input signal Pea and the output signal Peb of the experimental sensor Pe; Thus, the bleeding risk prediction function FXe of the characteristic area e is obtained, and then the risk interval is set for comparison, the bleeding prediction risk degree of the characteristic area e is evaluated and a risk warning signal is generated, and the characteristic area e of the human vascular distribution model is marked with corresponding risk.

[0006] Furthermore, the construction process of the comparison check sub-model is as follows: The output signals of the experimental sensor P and the calibration sensor Q are marked as Pb and Qb respectively; The frequency and amplitude of the output signal Pb of the experimental sensor P are marked as FREpb and AMPpb respectively; The frequency and amplitude of the output signal Qb of the calibration sensor Q are marked as FREqb and AMPqb respectively; The comparative evaluation process of frequency deviation is as follows: by comparing the output signal frequencies of the experimental sensor P and the calibration sensor Q at n0 identical time nodes, the frequency difference coefficient Ψfre is obtained: The comparative evaluation process of amplitude deviation is as follows: by comparing the output signal amplitudes of the experimental sensor P and the calibration sensor Q at n0 identical time nodes, the amplitude difference coefficient Ψamp is obtained: By combining the frequency difference coefficient Ψfre and the amplitude difference coefficient Ψamp, the signal accuracy evaluation index Zzq is obtained; The threshold of the signal accuracy evaluation index Zzq is set to MT0 and the threshold comparison is performed: when the signal accuracy evaluation index Zzq is lower than the threshold MT0, the acoustic wave information of the experimental sensor P is judged to be abnormal; otherwise, the acoustic wave information of the experimental sensor P is judged to be normal; When the acoustic wave information of the experimental sensor P is determined to be abnormal, the acoustic wave information of the experimental sensor P is refreshed and replaced by the acoustic wave information of the verification sensor Q, and a corresponding alarm is issued; when the acoustic wave information of the experimental sensor P is determined to be normal, no processing is performed.

[0007] Furthermore, the specific process of comprehensively analyzing the acoustic wave information of the pre-processed experimental sensor P is as follows: Compare the frequency FREpea of ​​the input signal Pea of ​​the experimental sensor Pe and the frequency FREpeb of the output signal Peb, and obtain the input frequency signal coefficients in turn. and the output frequency signal coefficient ; The amplitude AMPpea of ​​the input signal Pea of ​​the experimental sensor Pe and the input frequency signal coefficient Combined with the experimental sensor Pe, the energy index of the input signal Pea is obtained. , and the amplitude AMPpeb of the output signal Peb of the experimental sensor Pe and the output frequency signal coefficient Combined with the experimental sensor Pe, the energy index of the output signal Peb is obtained. ; Then, the energy index of the input signal Pea is and the energy index of the output signal Peb Combined, the bleeding risk prediction function FXe of the feature area e is obtained; The risk interval of the bleeding risk prediction function FXe of the feature region e is set, the bleeding prediction risk degree of the feature region e is evaluated by interval comparison, and a risk warning signal of a corresponding level is generated.

[0008] Furthermore, the specific evaluation process of the sensing performance of the surface acoustic wave filter is as follows: S3-1, evaluating the stability of the acoustic wave signal by analyzing the temporal changes of the acoustic wave signal; Set the data acquisition period Tc to regularly collect the bandwidth BWe and insertion loss ILe of the experimental sensor Pe, and obtain the attenuation evaluation coefficient Xsh by combining the bandwidth BWe and the insertion loss ILe, so as to evaluate the stability of the acoustic wave signal; S3-2, evaluating the real-time performance of the acoustic signal by monitoring the response delay of the acoustic signal; By collecting the delay time Ty between the input signal Pea and the output signal Peb of the experimental sensor Pe, the response delay coefficient Xys is obtained, thereby evaluating the real-time performance of the acoustic wave signal; S3-3, and then comprehensively evaluate the sensing performance of the SAW filter; By combining the attenuation evaluation coefficient Xsh and the response delay coefficient Xys of the experimental sensor Pe, and regularly measuring and calculating the sensing performance index SP of the surface acoustic wave filter according to the data acquisition period Tc, the sensing performance of the surface acoustic wave filter is evaluated.

[0009] Furthermore, the specific process of monitoring and analyzing interference factors is as follows: S4-1, regularly monitoring and collecting interference information of the surface acoustic wave filter through a data collection period Tc, the interference information including electric field intensity Be, magnetic field intensity He, mechanical vibration frequency Ze and ambient temperature Fe; S4-101, construct and analyze the parameter analysis matrix V through the electric field intensity Be, the magnetic field intensity He, the mechanical vibration frequency Ze and the ambient temperature Fe and the sensor performance index SP, obtain the corresponding influence functions, and mark them as the electric field intensity influence function F1, the magnetic field intensity influence function F2, the mechanical vibration influence function F3 and the ambient temperature influence function F4 respectively; S4-102, through the electric field strength influence function F1, magnetic field strength influence function F2, mechanical vibration influence function F3 and ambient temperature influence function F4, obtain the weight coefficients of electric field strength Be, magnetic field strength He, mechanical vibration frequency Ze and ambient temperature Fe for the sensing performance index SP and mark them as υ1, υ2, υ3, υ4 respectively, so as to evaluate the influence of electric field strength Be, magnetic field strength He, mechanical vibration frequency Ze and ambient temperature Fe on the sensing performance index SP, and then obtain the comprehensive influence coefficient ZH of the sensing performance of the surface acoustic wave filter.

[0010] Furthermore, the specific process of risk prediction for the sensing performance of the surface acoustic wave filter is as follows: S4-2, constructing a variation curve sv between the sensing performance index SP and the data acquisition period Tc through the v5 column vector of the parameter analysis matrix V; S4-201, extracting n3 points of the change curve sv and their coordinates, obtaining the slope value Ki of any point i, obtaining the curve increase rate value Kv of the change curve sv by measuring the average value of the slope values ​​of the n3 points of the change curve sv, and then measuring the standard deviation of the slope values ​​of the n3 points to obtain the increase rate fluctuation coefficient σv of the change curve sv; S4-202, setting a risk prediction period Tr, and obtaining a risk prediction index RISK of the sensor performance by combining the curve increase rate value Kv of the change curve sv, the increase rate fluctuation coefficient σv, and the sensor performance comprehensive influence coefficient ZH; S4-203, set the warning interval of the risk prediction index RISK of the sensor performance, evaluate the aging degree of the surface acoustic wave filter device and issue corresponding warning prompt signals, so as to carry out corresponding device optimization management.

[0011] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: The present invention constructs a human blood vessel distribution model through an acoustic wave information detection module, detects and collects the acoustic wave information of waste liquid in the characteristic area of ​​the human body during the blood purification process, and then designs a blood leakage monitoring and early warning model through an early warning model design module, evaluates the risk degree of blood leakage prediction, and performs corresponding human blood vessel blood leakage risk marking, thereby realizing the blood leakage risk monitoring of different puncture points of hemodialysis, and ensuring the pertinence and accuracy of blood leakage monitoring; The present invention comprehensively evaluates the sensing performance of the surface acoustic wave filter from the stability and real-time performance of the acoustic wave signal through a sensing performance evaluation module, and then introduces interference factors through a remote response optimization module to analyze the influence of the filter sensing performance, thereby predicting the risk of the sensing performance of the surface acoustic wave filter under long-term use, evaluating the aging degree of the surface acoustic wave filter device and performing device optimization management, thereby ensuring the continuous stability of the sensing performance of the surface acoustic wave filter, and finally realizing the targeted, highly accurate and stable blood leakage monitoring in the blood purification system by using the surface acoustic wave. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 The module connection diagram of the present invention is shown; Figure 2 A schematic diagram of the process steps of the present invention is shown. DETAILED DESCRIPTION

[0013] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.

[0014] Embodiment 1: like Figure 1-Figure 2 As shown, a device for realizing blood leakage monitoring by using surface acoustic waves in a blood purification system comprises an acoustic wave information detection module, an early warning model design module, a sensor performance evaluation module and a remote response optimization module, wherein the acoustic wave information detection module, the early warning model design module, the sensor performance evaluation module and the remote response optimization module are communicatively connected; S1, the acoustic wave information detection module detects and collects the acoustic wave information of the waste liquid: constructs a human blood vessel distribution model and marks N0 characteristic areas, detects the waste liquid in the blood purification process through a surface acoustic wave filter, detects the waste liquid in N0 characteristic areas in real time, and collects the corresponding acoustic wave information; The sound wave information includes the frequency and amplitude of the sound wave signal, and the sound wave signal includes an input signal and an output signal; The characteristic area refers to the area where the hemodialysis puncture point is located, such as the subclavian artery, femoral artery, radial artery, internal jugular vein and femoral vein. The blood flow velocity and thrombosis situation of the blood vessels in the areas where different puncture points are located are different, so the risk of bleeding is also different.

[0015] S2, the early warning model design module designs a blood leakage monitoring early warning model: the sound wave information of N0 feature areas is analyzed and processed to design a blood leakage monitoring early warning model; S2-1, construct a comparison check sub-model to pre-process the sound wave information. The specific process is as follows: The surface acoustic wave filter for real-time detection of waste liquid is marked as an experimental sensor P, and a redundant surface acoustic wave filter is set to perform regular verification of the waste liquid, the regular verification period is marked as Tj, and the redundant surface acoustic wave filter is marked as a verification sensor Q. By comparing the experimental sensor P with the verification sensor Q, the data of the experimental sensor P is verified; The input signals of the experimental sensor P and the verification sensor Q are set to be the same. After the experimental sensor P and the verification sensor Q monitor the waste liquid, the corresponding output signals are obtained respectively. By comparing the acoustic wave information of the two sets of output signals at the same time node, the frequency deviation and amplitude deviation between the experimental sensor P and the verification sensor Q are evaluated, thereby evaluating the accuracy of the acoustic wave signal of the experimental sensor P; The construction process of the comparison and verification sub-model is as follows: the output signals of the experimental sensor P and the verification sensor Q are marked as Pb and Qb respectively; Among them, frequency refers to the signal frequency of the sound wave signal; amplitude refers to the signal amplitude of the sound wave signal; The frequency and amplitude of the output signal Pb of the experimental sensor P are marked as FREpb and AMPpb respectively; The frequency and amplitude of the output signal Qb of the calibration sensor Q are marked as FREqb and AMPqb respectively; The comparative evaluation process of frequency deviation is as follows: by comparing the output signal frequencies of the experimental sensor P and the calibration sensor Q at n0 identical time nodes, the frequency difference coefficient Ψfre is obtained: ; The comparative evaluation process of amplitude deviation is as follows: by comparing the output signal amplitudes of the experimental sensor P and the calibration sensor Q at n0 identical time nodes, the amplitude difference coefficient Ψamp is obtained: ; By combining the frequency difference coefficient Ψfre and the amplitude difference coefficient Ψamp, the signal accuracy evaluation index Zzq is obtained: ; Among them, α1 and α2 are weight factor coefficients of the frequency difference coefficient Ψfre and the amplitude difference coefficient Ψamp, respectively, and α1 and α2 are both greater than 0. The weight factor coefficients are preset and obtained after a large amount of experimental data is measured. When the frequency difference coefficient Ψfre and the amplitude difference coefficient Ψamp are higher, the signal accuracy evaluation index Zzq is lower, indicating that the degree of deviation between the experimental sensor P and the verification sensor Q is higher, and the accuracy of the acoustic wave signal of the experimental sensor P is evaluated to be lower; since the experimental sensor P is in a real-time monitoring state, and the device loss caused by long-term use will cause the data monitoring accuracy of the experimental sensor P to continue to decrease, while the verification sensor Q is in a timed verification state, and the use cycle and device loss are relatively low compared to the experimental sensor P, therefore, by comparing the degree of deviation between the experimental sensor P and the verification sensor Q, the accuracy of the acoustic wave signal of the experimental sensor P is evaluated; The threshold of the signal accuracy evaluation index Zzq is set to MT0 and the threshold comparison is performed: when the signal accuracy evaluation index Zzq is lower than the threshold MT0, the acoustic wave information of the experimental sensor P is judged to be abnormal; otherwise, the acoustic wave information of the experimental sensor P is judged to be normal; When the acoustic wave information of the experimental sensor P is determined to be abnormal, the acoustic wave information of the experimental sensor P is refreshed and replaced by the acoustic wave information of the verification sensor Q, and a corresponding alarm is issued; When the acoustic wave information of the experimental sensor P is judged to be normal, no processing is performed; Preprocessing the acoustic wave information by comparing the check sub-model; S2-2, then comprehensively analyze the acoustic wave information of the experimental sensor P after preprocessing, the specific process is as follows: Mark any characteristic region as e, mark the surface acoustic wave filter of the characteristic region e as the experimental sensor Pe, and compare the input signal Pea and the output signal Peb of the experimental sensor Pe; Compare the frequency FREpea of ​​the input signal Pea of ​​the experimental sensor Pe and the frequency FREpeb of the output signal Peb; From the Fourier transform: The analysis yields: ; in, is the time domain signal, is the frequency domain signal, f is the frequency; The input signal Pea and the output signal Peb are transformed by Fourier transform to obtain the input frequency signal coefficients in turn. and the output frequency signal coefficient ; Input frequency signal coefficient: , when the input frequency domain signal coefficient The higher the input signal Pea is, the higher the frequency The stronger the place; Output frequency signal coefficient: , when the output frequency domain signal coefficient The higher the value, the higher the output signal Peb. The stronger the place; The amplitude AMPpea of ​​the input signal Pea of ​​the experimental sensor Pe and the input frequency signal coefficient Combined with the experimental sensor Pe, the energy index of the input signal Pea is obtained. , and the amplitude AMPpeb of the output signal Peb of the experimental sensor Pe and the output frequency signal coefficient Combined with the experimental sensor Pe, the energy index of the output signal Peb is obtained. ; Then, the energy index of the input signal Pea is and the energy index of the output signal Peb Combined, the bleeding risk prediction function FXe of the feature area e is obtained: ; Among them, n1 time nodes are set to collect signal amplitude, β1 and β2 are energy indexes respectively. and energy index The weight factor coefficient, and β1 and β2 are both greater than 0; when the energy index of the input signal Pea and the energy index of the output signal Peb The smaller the difference, the lower the loss of the acoustic signal. When the waste liquid contains blood test solids, the loss of the acoustic signal will increase. Therefore, when the energy index of the input signal Pea is and the energy index of the output signal Peb If the difference exceeds the preset risk interval, it is determined that the feature area has a risk of bleeding. When the bleeding risk prediction function FXe is higher, the predicted risk of bleeding in the feature area e is higher. Setting the risk interval of the bleeding risk prediction function FXe of the characteristic region e, evaluating the bleeding prediction risk degree of the characteristic region e by comparing the intervals, generating a risk warning signal of a corresponding level, and then marking the corresponding risk on the characteristic region e of the human vascular distribution model; For example, according to the degree of predicted risk of bleeding from light to heavy, the level of risk warning signal is set from small to large, and risk signs of different colors are set from light to dark, and risk marking is performed on characteristic areas to achieve a comprehensive analysis of the human blood vessel distribution model, and targeted bleeding monitoring is performed on different characteristic parts. Since the degree of risk marking in characteristic areas is different, the predicted probability of bleeding risk at different puncture points is different, thereby achieving targeted prediction of bleeding risk in characteristic areas.

[0016] S3, the sensing performance evaluation module evaluates the sensing performance of the surface acoustic wave filter: by analyzing the timing changes of the acoustic wave signal, the stability of the acoustic wave signal is evaluated; and the response delay of the acoustic wave signal is monitored to evaluate the real-time performance of the acoustic wave signal; and then the sensing performance of the surface acoustic wave filter is comprehensively evaluated; S3-1, evaluating the stability of the acoustic wave signal by analyzing the temporal changes of the acoustic wave signal; Set the data acquisition period Tc to regularly collect the bandwidth BWe and insertion loss ILe of the experimental sensor Pe. By combining the bandwidth BWe and the insertion loss ILe, the attenuation evaluation coefficient Xsh is obtained: ; Among them, ‌Bandwidth‌ refers to the frequency range that the filter can pass; ‌Insertion loss‌ refers to the ratio of the power lost after the signal passes through the filter to the input power; ILe0 and BWe0 are the data values ​​of the bandwidth BWe and insertion loss ILe of the experimental sensor Pe corresponding to the current time node, that is, the performance index parameters of the current surface acoustic wave filter after the acoustic wave signal is detected for a certain period of time; BWp refers to the maximum frequency of the input / output signal detected. When the surface acoustic wave filter can pass the maximum frequency of the input / output signal, then When it is greater than 0, it means the attenuation is smaller; when the bandwidth BWe of the experimental sensor Pe is higher and the insertion loss ILe is lower, the attenuation evaluation coefficient Xsh is lower, indicating that the bandwidth attenuation is low and the insertion loss enhancement is small. The comprehensive result makes the attenuation evaluation degree low, which means that the performance of the filter is better and the stability of the evaluated acoustic wave signal is better; S3-2, evaluating the real-time performance of the acoustic signal by monitoring the response delay of the acoustic signal; The acoustic wave signals are transmitted and received by two groups of interdigital transducers respectively, so as to collect the delay time Ty between the input signal Pea and the output signal Peb of the experimental sensor Pe, and obtain the response delay coefficient Xys: ; Among them, by collecting the delay time between the n2 groups of input signals Pea and output signals Peb, when the delay time Ty is higher, the response delay coefficient Xys is higher, the more serious the evaluation response delay is, the worse the real-time performance of the acoustic signal is; S3-3, and then comprehensively evaluate the sensing performance of the SAW filter; By combining the attenuation evaluation coefficient Xsh and the response delay coefficient Xys of the experimental sensor Pe, and regularly measuring and calculating the sensing performance index SP of the surface acoustic wave filter according to the data acquisition period Tc, the sensing performance of the surface acoustic wave filter is evaluated; The sensing performance index SP of the surface acoustic wave filter: ; Among them, μ1 and μ2 are weight factor coefficients of the attenuation evaluation coefficient Xsh and the response delay coefficient Xys, respectively, and both μ1 and μ2 are greater than 0; when the attenuation evaluation coefficient Xsh and the response delay coefficient Xys are lower, the sensing performance index SP of the surface acoustic wave filter is higher, and the sensing performance of the surface acoustic wave filter is evaluated to be better.

[0017] S4, remote response optimization module remote blood leakage monitoring and device optimization management: through risk prediction of the sensing performance of the surface acoustic wave filter, the aging degree of the device is evaluated and corresponding early warning prompts are issued, so as to optimize the device management; S4-1, regularly monitoring and collecting interference information of the surface acoustic wave filter through a data collection period Tc, the interference information including electric field intensity Be, magnetic field intensity He, mechanical vibration frequency Ze and ambient temperature Fe; Among them, remote blood leakage monitoring is to transmit signal data through the Internet and remotely monitor interference information, which is collected through existing sensors, such as electric field meters, flux meters, vibration sensors and thermometers; The parameter analysis matrix V is constructed by the electric field intensity Be, magnetic field intensity He, mechanical vibration frequency Ze, ambient temperature Fe and sensor performance index SP: ; The electric field intensity Be, magnetic field intensity He, mechanical vibration frequency Ze, ambient temperature Fe and sensor performance index SP are respectively used as 5 column vectors of parameter analysis matrix V; the interference information corresponding to m1 data acquisition cycles Tc is collected and respectively used as m1 row vectors of parameter analysis matrix V; Through the column vector of the electric field intensity Be and the column vector of the sensing performance index SP, a variation curve s1 between the electric field intensity Be and the sensing performance index SP is constructed, and the electric field intensity influence function F1 is generated by fitting: ; Substitute the electric field strength Be into the electric field strength influence function F1 to obtain the corresponding sensing performance index SP; Through the column vector of magnetic field intensity He and the column vector of sensing performance index SP, a change curve s2 between magnetic field intensity He and sensing performance index SP is constructed, and the magnetic field intensity influence function F2 is generated by fitting: ; Substitute the magnetic field intensity He into the magnetic field intensity influence function F2 to obtain the corresponding sensing performance index SP; Through the column vector of the mechanical vibration frequency Ze and the column vector of the sensor performance index SP, a change curve s3 between the mechanical vibration frequency Ze and the sensor performance index SP is constructed, and the mechanical vibration influence function F3 is generated by fitting: ; Substitute the mechanical vibration frequency Ze into the electric field strength influence function F3 to obtain the corresponding sensing performance index SP; Through the column vector of the ambient temperature Fe and the column vector of the sensor performance index SP, a change curve s4 between the ambient temperature Fe and the sensor performance index SP is constructed, and the ambient temperature influence function F4 is generated by fitting: ; Substitute the ambient temperature Fe into the ambient temperature influence function F4 to obtain the corresponding sensor performance index SP; From this we can get, , then υ1, υ2, υ3, υ4 are the weight coefficients of the electric field intensity Be, magnetic field intensity He, mechanical vibration frequency Ze and ambient temperature Fe respectively, and υ5 is the sensor performance index SP of the current data acquisition period Tc, that is, 4 times; in, , , , ; Through the electric field intensity influence function F1, the magnetic field intensity influence function F2, the mechanical vibration influence function F3 and the ambient temperature influence function F4, the weight coefficients υ1, υ2, υ3 and υ4 are obtained in turn, so as to evaluate the influence of the electric field intensity Be, the magnetic field intensity He, the mechanical vibration frequency Ze and the ambient temperature Fe on the sensing performance index SP; By introducing interference factors such as electromagnetic intensity, mechanical vibration and ambient temperature, the influence of interference factors on the electronic components of the surface acoustic wave filter under long-term use is analyzed; S4-2, by using the v5 column vector of the parameter analysis matrix V, with the sensing performance index SP as the ordinate and the data acquisition period Tc as the abscissa, construct a change curve sv between the sensing performance index SP and the data acquisition period Tc, and by analyzing the growth fluctuation of the change curve sv, perform risk prediction on the sensing performance of the surface acoustic wave filter; Extract n3 points and their coordinates of the change curve sv, obtain the slope value Ki of any point i, mark any point and its coordinates of the change curve sv as i (Xi, Yi), mark the point closest to point i and its coordinates as g (Xg, Yg), and obtain the slope value Ki of point i of the change curve sv by combining i (Xi, Yi) and g (Xg, Yg): ; By measuring the average value of the slope values ​​of n3 points of the change curve sv, the curve increase rate value Kv of the change curve sv is obtained: , and then calculate the standard deviation of the slope values ​​of n3 points to obtain the rate fluctuation coefficient σv of the change curve sv: ; Set the risk prediction period Tr, and obtain the risk prediction index RISK of the sensor performance by combining the curve increase rate value Kv and the increase rate fluctuation coefficient σv of the change curve sv and the comprehensive influence coefficient ZH of the sensor performance: ; Among them, when the comprehensive influence coefficient ZH of sensing performance is higher, it means that the interference factor has a higher influence on the sensing performance, which will make the risk level of the sensing performance higher; Ym is the ordinate of the end point of the change curve sv. When Ym is lower, it means that the sensing performance of the surface acoustic wave filter at the current time node is worse; Tr refers to the time length from the current time node to the predicted time node, that is, the risk prediction period. When the risk prediction period Tr is higher, it means that the possibility of uncertain factors occurring is higher, which will make the risk level of the sensing performance higher; ρ1 and ρ2 are the weight factor coefficients of the rate fluctuation coefficient σv and the curve rate value Kv, respectively, and both ρ1 and ρ2 are greater than 0; when the rate fluctuation coefficient σv of the change curve sv is higher, it means that the change of the sensor performance is more unstable and the risk level is higher; when the curve rate value Kv is negative and the lower it is, it means that the downward trend of the sensor performance is more obvious and the risk level is higher, then the risk prediction index RISK is higher; conversely, when the curve rate value Kv is positive and the higher it is, it means that the upward trend of the sensor performance is more obvious and the risk prediction index RISK is lower; Set the warning interval of the risk prediction index RISK of the sensor performance, evaluate the aging degree of the surface acoustic wave filter device and issue corresponding warning prompt signals, so as to carry out corresponding device optimization management according to the warning prompt signals; Among them, device optimization management mainly involves corresponding processing of surface acoustic wave filters, such as cleaning, calibration or replacement operations, to avoid risks such as device aging and attenuation caused by long-term use and interference factors, and to optimize filter risks in advance, thereby ensuring the continuous stability of the sensing performance of the surface acoustic wave filter and achieving stable and accurate blood leakage monitoring.

[0018] In summary, the present invention constructs a human blood vessel distribution model through an acoustic wave information detection module, detects and collects the acoustic wave information of waste liquid in the characteristic area of ​​the human body during the blood purification process, and then designs a blood leakage monitoring and early warning model through an early warning model design module, evaluates the risk degree of blood leakage prediction and performs corresponding human blood vessel blood leakage risk marking. By performing targeted analysis and risk marking on specific areas, the blood leakage risk monitoring of different puncture points of hemodialysis is realized, ensuring the pertinence and accuracy of blood leakage monitoring; The present invention comprehensively evaluates the sensing performance of the surface acoustic wave filter from the two perspectives of the stability and real-time performance of the acoustic wave signal through a sensing performance evaluation module, and then performs remote blood leakage monitoring through a remote response optimization module, and introduces interference factors to analyze the influence of the filter sensing performance, so as to perform risk prediction on the sensing performance of the surface acoustic wave filter under long-term use, evaluate the device aging degree of the surface acoustic wave filter and perform device optimization management, so as to ensure that the sensing performance of the surface acoustic wave filter is continuously stable, and finally realize the blood leakage monitoring with strong pertinence, high accuracy and good stability in the blood purification system by using surface acoustic waves.

[0019] The size of the interval and threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by technical personnel in this field for each group of sample data; as long as it does not affect the proportional relationship between the parameter and the quantized value.

[0020] The above formulas are all dimensionless and numerical calculations. The formula is a formula obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formula are set by technicians in this field according to actual conditions. The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A device for monitoring blood leakage using surface acoustic waves in a blood purification system, characterized in that: It includes an acoustic information detection module, an early warning model design module, a sensor performance evaluation module and a remote response optimization module, wherein the acoustic information detection module, the early warning model design module, the sensor performance evaluation module and the remote response optimization module are communicatively connected; The acoustic wave information detection module is used to detect the acoustic wave information of the collected waste liquid: construct a human blood vessel distribution model and mark N0 characteristic areas, detect the waste liquid in the blood purification process through the surface acoustic wave filter, detect the waste liquid in N0 characteristic areas in real time, and collect the corresponding acoustic wave information; The early warning model design module is used to design a blood leakage monitoring early warning model: by analyzing and processing the acoustic wave information of N0 characteristic areas, a blood leakage monitoring early warning model is designed, the degree of blood leakage prediction risk in the characteristic areas is evaluated, and corresponding risk warning signals are generated, and then risk marking is performed on the characteristic areas of the human blood vessel distribution model; The sensing performance evaluation module is used to evaluate the sensing performance of the surface acoustic wave filter: by analyzing the timing changes of the acoustic wave signal, the stability of the acoustic wave signal is evaluated, and by monitoring the response delay of the acoustic wave signal, the real-time performance of the acoustic wave signal is evaluated. Then, the sensing performance of the surface acoustic wave filter is comprehensively evaluated through the stability and real-time performance of the acoustic wave signal. The remote response optimization module is used for remote blood leakage monitoring and device optimization management: by monitoring and analyzing the impact of interference factors on the electronic components of the surface acoustic wave filter, the risk of the sensing performance of the surface acoustic wave filter is predicted, the aging degree of the surface acoustic wave filter device is evaluated and corresponding early warning prompts are issued, thereby optimizing the device management.

2. The device for monitoring blood leakage using surface acoustic waves in a blood purification system according to claim 1, characterized in that: The specific design process of the blood leakage monitoring and early warning model is as follows: S2-1, construct a comparison check sub-model to pre-process the sound wave information: The surface acoustic wave filter for real-time detection of waste liquid is marked as an experimental sensor P, and a redundant surface acoustic wave filter is set to perform regular verification of the waste liquid, the regular verification period is marked as Tj, and the redundant surface acoustic wave filter is marked as a verification sensor Q. By comparing the experimental sensor P with the verification sensor Q, the data of the experimental sensor P is verified; The sound wave information includes the frequency and amplitude of the sound wave signal, and the sound wave signal includes an input signal and an output signal; The input signals of the experimental sensor P and the verification sensor Q are set to be the same. After the experimental sensor P and the verification sensor Q monitor the waste liquid, the corresponding output signals are obtained respectively. By comparing the acoustic wave information of the two sets of output signals at the same time node, the frequency deviation and amplitude deviation between the experimental sensor P and the verification sensor Q are evaluated, thereby evaluating the accuracy of the acoustic wave signal of the experimental sensor P, determining the abnormality of the acoustic wave information, and performing corresponding verification processing; S2-2, then conduct a comprehensive analysis on the acoustic wave information of the pre-processed experimental sensor P: Mark any characteristic region as e, mark the surface acoustic wave filter of the characteristic region e as the experimental sensor Pe, and compare the input signal Pea and the output signal Peb of the experimental sensor Pe; Thus, the bleeding risk prediction function FXe of the characteristic area e is obtained, and then the risk interval is set for comparison, the bleeding prediction risk degree of the characteristic area e is evaluated and a risk warning signal is generated, and the characteristic area e of the human vascular distribution model is marked with corresponding risk.

3. The device for monitoring blood leakage using surface acoustic waves in a blood purification system according to claim 2, characterized in that: The construction process of the comparison check sub-model is as follows: The output signals of the experimental sensor P and the calibration sensor Q are marked as Pb and Qb respectively; The frequency and amplitude of the output signal Pb of the experimental sensor P are marked as FREpb and AMPpb respectively; The frequency and amplitude of the output signal Qb of the calibration sensor Q are marked as FREqb and AMPqb respectively; The comparative evaluation process of frequency deviation is as follows: by comparing the output signal frequencies of the experimental sensor P and the calibration sensor Q at n0 identical time nodes, the frequency difference coefficient Ψfre is obtained: The comparative evaluation process of amplitude deviation is as follows: by comparing the output signal amplitudes of the experimental sensor P and the calibration sensor Q at n0 identical time nodes, the amplitude difference coefficient Ψamp is obtained: By combining the frequency difference coefficient Ψfre and the amplitude difference coefficient Ψamp, the signal accuracy evaluation index Zzq is obtained; The threshold of the signal accuracy evaluation index Zzq is set to MT0 and the threshold comparison is performed: when the signal accuracy evaluation index Zzq is lower than the threshold MT0, the acoustic wave information of the experimental sensor P is judged to be abnormal; otherwise, the acoustic wave information of the experimental sensor P is judged to be normal; When the acoustic wave information of the experimental sensor P is determined to be abnormal, the acoustic wave information of the experimental sensor P is refreshed and replaced by the acoustic wave information of the verification sensor Q, and a corresponding alarm is issued; when the acoustic wave information of the experimental sensor P is determined to be normal, no processing is performed.

4. The device for monitoring blood leakage using surface acoustic waves in a blood purification system according to claim 3, characterized in that: The specific process of comprehensive analysis of the acoustic wave information of the pre-processed experimental sensor P is as follows: Compare the frequency FREpea of ​​the input signal Pea of ​​the experimental sensor Pe and the frequency FREpeb of the output signal Peb, and obtain the input frequency signal coefficients in turn. and the output frequency signal coefficient ; The amplitude AMPpea of ​​the input signal Pea of ​​the experimental sensor Pe and the input frequency signal coefficient Combined with the experimental sensor Pe, the energy index of the input signal Pea is obtained. , and the amplitude AMPpeb of the output signal Peb of the experimental sensor Pe and the output frequency signal coefficient Combined with the experimental sensor Pe, the energy index of the output signal Peb is obtained. ; Then, the energy index of the input signal Pea is and the energy index of the output signal Peb Combined, the bleeding risk prediction function FXe of the feature area e is obtained; The risk interval of the bleeding risk prediction function FXe of the feature region e is set, the bleeding prediction risk degree of the feature region e is evaluated by interval comparison, and a risk warning signal of a corresponding level is generated.

5. The device for monitoring blood leakage using surface acoustic waves in a blood purification system according to claim 4, characterized in that: The specific evaluation process of the sensing performance of the surface acoustic wave filter is as follows: S3-1, evaluating the stability of the acoustic wave signal by analyzing the temporal changes of the acoustic wave signal; Set the data acquisition period Tc to regularly collect the bandwidth BWe and insertion loss ILe of the experimental sensor Pe, and obtain the attenuation evaluation coefficient Xsh by combining the bandwidth BWe and the insertion loss ILe, so as to evaluate the stability of the acoustic wave signal; S3-2, evaluating the real-time performance of the acoustic signal by monitoring the response delay of the acoustic signal; By collecting the delay time Ty between the input signal Pea and the output signal Peb of the experimental sensor Pe, the response delay coefficient Xys is obtained, thereby evaluating the real-time performance of the acoustic wave signal; S3-3, and then comprehensively evaluate the sensing performance of the SAW filter; By combining the attenuation evaluation coefficient Xsh and the response delay coefficient Xys of the experimental sensor Pe, and regularly measuring and calculating the sensing performance index SP of the surface acoustic wave filter according to the data acquisition period Tc, the sensing performance of the surface acoustic wave filter is evaluated.

6. The device for monitoring blood leakage using surface acoustic waves in a blood purification system according to claim 5, characterized in that: The specific process of monitoring and analyzing interference factors is as follows: S4-1, regularly monitoring and collecting interference information of the surface acoustic wave filter through a data collection period Tc, the interference information including electric field intensity Be, magnetic field intensity He, mechanical vibration frequency Ze and ambient temperature Fe; S4-101, construct and analyze the parameter analysis matrix V through the electric field intensity Be, the magnetic field intensity He, the mechanical vibration frequency Ze and the ambient temperature Fe and the sensor performance index SP, obtain the corresponding influence functions, and mark them as the electric field intensity influence function F1, the magnetic field intensity influence function F2, the mechanical vibration influence function F3 and the ambient temperature influence function F4 respectively; S4-102, through the electric field strength influence function F1, magnetic field strength influence function F2, mechanical vibration influence function F3 and ambient temperature influence function F4, obtain the weight coefficients of electric field strength Be, magnetic field strength He, mechanical vibration frequency Ze and ambient temperature Fe for the sensing performance index SP and mark them as υ1, υ2, υ3, υ4 respectively, so as to evaluate the influence of electric field strength Be, magnetic field strength He, mechanical vibration frequency Ze and ambient temperature Fe on the sensing performance index SP, and then obtain the comprehensive influence coefficient ZH of the sensing performance of the surface acoustic wave filter.

7. The device for monitoring blood leakage using surface acoustic waves in a blood purification system according to claim 6, characterized in that: The specific process of risk prediction for the sensing performance of the surface acoustic wave filter is as follows: S4-2, constructing a variation curve sv between the sensing performance index SP and the data acquisition period Tc through the v5 column vector of the parameter analysis matrix V; S4-201, extracting n3 points of the change curve sv and their coordinates, obtaining the slope value Ki of any point i, obtaining the curve increase rate value Kv of the change curve sv by measuring the average value of the slope values ​​of the n3 points of the change curve sv, and then measuring the standard deviation of the slope values ​​of the n3 points to obtain the increase rate fluctuation coefficient σv of the change curve sv; S4-202, setting a risk prediction period Tr, and obtaining a risk prediction index RISK of the sensor performance by combining the curve increase rate value Kv of the change curve sv, the increase rate fluctuation coefficient σv, and the sensor performance comprehensive influence coefficient ZH; S4-203, set the warning interval of the risk prediction index RISK of the sensor performance, evaluate the aging degree of the surface acoustic wave filter device and issue corresponding warning prompt signals, so as to carry out corresponding device optimization management.

Images