CN115969464B - Method and system for predicting thrombolysis effect of piezoelectric impedance based on regression of support vector machine - Google Patents
Method and system for predicting thrombolysis effect of piezoelectric impedance based on regression of support vector machine Download PDFInfo
- Publication number
- CN115969464B CN115969464B CN202211678107.2A CN202211678107A CN115969464B CN 115969464 B CN115969464 B CN 115969464B CN 202211678107 A CN202211678107 A CN 202211678107A CN 115969464 B CN115969464 B CN 115969464B
- Authority
- CN
- China
- Prior art keywords
- support vector
- vector machine
- target
- thrombolytic
- thrombus
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000002537 thrombolytic effect Effects 0.000 title claims abstract description 106
- 238000012706 support-vector machine Methods 0.000 title claims abstract description 79
- 230000000694 effects Effects 0.000 title claims abstract description 43
- 238000000034 method Methods 0.000 title claims abstract description 33
- 208000007536 Thrombosis Diseases 0.000 claims abstract description 57
- 241000282461 Canis lupus Species 0.000 claims abstract description 52
- 238000012360 testing method Methods 0.000 claims abstract description 47
- 238000012549 training Methods 0.000 claims abstract description 41
- 239000003146 anticoagulant agent Substances 0.000 claims abstract description 35
- 239000008280 blood Substances 0.000 claims abstract description 29
- 210000004369 blood Anatomy 0.000 claims abstract description 29
- 238000012545 processing Methods 0.000 claims abstract description 20
- 230000006870 function Effects 0.000 claims description 62
- 241000282421 Canidae Species 0.000 claims description 41
- 238000003756 stirring Methods 0.000 claims description 14
- 238000005457 optimization Methods 0.000 claims description 9
- 238000010606 normalization Methods 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 5
- 230000036772 blood pressure Effects 0.000 claims description 4
- 230000001732 thrombotic effect Effects 0.000 claims 1
- 239000000243 solution Substances 0.000 description 29
- 239000003527 fibrinolytic agent Substances 0.000 description 6
- 239000013078 crystal Substances 0.000 description 5
- 208000006011 Stroke Diseases 0.000 description 4
- 210000004204 blood vessel Anatomy 0.000 description 4
- 239000003814 drug Substances 0.000 description 4
- 229960000103 thrombolytic agent Drugs 0.000 description 4
- 230000015572 biosynthetic process Effects 0.000 description 3
- 230000002490 cerebral effect Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 206010008190 Cerebrovascular accident Diseases 0.000 description 2
- 208000032382 Ischaemic stroke Diseases 0.000 description 2
- 210000004556 brain Anatomy 0.000 description 2
- 208000029028 brain injury Diseases 0.000 description 2
- 230000006378 damage Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 229940079593 drug Drugs 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 208000024891 symptom Diseases 0.000 description 2
- 230000001225 therapeutic effect Effects 0.000 description 2
- 206010008132 Cerebral thrombosis Diseases 0.000 description 1
- 206010018985 Haemorrhage intracranial Diseases 0.000 description 1
- 208000008574 Intracranial Hemorrhages Diseases 0.000 description 1
- 201000001429 Intracranial Thrombosis Diseases 0.000 description 1
- 102000003990 Urokinase-type plasminogen activator Human genes 0.000 description 1
- 108090000435 Urokinase-type plasminogen activator Proteins 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 238000004090 dissolution Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000002347 injection Methods 0.000 description 1
- 239000007924 injection Substances 0.000 description 1
- 230000002427 irreversible effect Effects 0.000 description 1
- 230000000302 ischemic effect Effects 0.000 description 1
- 208000011977 language disease Diseases 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 235000015097 nutrients Nutrition 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 229960005356 urokinase Drugs 0.000 description 1
- 230000002792 vascular Effects 0.000 description 1
Landscapes
- External Artificial Organs (AREA)
Abstract
The invention relates to a piezoelectric impedance thrombolysis effect prediction method and a system based on support vector machine regression, comprising the following steps: testing the piezoresistance data corresponding to thrombolytic solutions with different concentrations by using a thrombus vibration actuator; carrying out standardized processing on the piezoresistor resistance data to obtain a training set and a testing set; taking the training set and the testing set as characteristic parameters, inputting the characteristic parameters into a support vector machine regression model optimized by a gray wolf algorithm for training and testing, and obtaining a trained support vector machine regression model; predicting the blood concentration of the target to be detected based on the trained support vector machine regression model and the target piezoresistance data; and predicting the thrombolysis effect of the target to be tested based on the blood concentration. The invention provides a piezoelectric impedance thrombolysis effect prediction method adopting support vector machine regression for the first time, and the technical problems of low reliability and low accuracy in the prior art are alleviated.
Description
Technical Field
The invention relates to the technical field of thrombolysis effect prediction, in particular to a method and a system for predicting a thrombolysis effect of piezoelectric impedance based on support vector machine regression.
Background
According to global disease burden study (Global burden of disease study, GBD) data, cerebral apoplexy is a common cause of death disability in the first three places in the world, and is also a primary cause of death disability in China. Tens of millions of people worldwide suffer from stroke each year, a significant portion of which is ischemic stroke. The main cause of ischemic cerebral apoplexy is that cerebral thrombosis blocks blood vessels, reduces or cuts off nutrient supply to the brain, and causes brain injury, thereby causing language disorder, limb disorder and even death. In many cases, stroke occurs suddenly, causing irreversible damage to the brain after a few hours. Therefore, the thrombus should be cleared and dissolved as soon as possible to reopen the blocked blood vessel, avoiding brain injury.
There are two main methods for treating ischemic stroke internationally: thrombolysis and mechanical thrombolysis. Among them, the use of injection thrombolytic agents such as rt-PA, urokinase and other drugs for drug thrombolysis is a worldwide accepted therapeutic approach. However, there is a therapeutic window using thrombolysis, typically 3-4.5 hours after symptoms occur. Because patients spend a great deal of time delivering medicine and confirming the etiology after symptoms appear, most patients do not have a treatment with thrombolytic drugs within the time window. Meanwhile, the thrombolysis has certain disadvantages, such as long vascular recanalization time and low recanalization rate, and intracranial hemorrhage is easy to occur when a large amount of thrombolytic agents are used. Therefore, mechanical intervention thrombus removing systems appear at home and abroad, and thrombus is grabbed at the internal position of a blood vessel through a self-expansion spiral ring or a bracket mechanism and taken out of the body. Compared with the medicine thrombolysis, the mechanical thrombolysis system has the advantages of longer time window, higher recanalization rate, faster recanalization time and the like. But due to the characteristics of self-expanding stent structure it is prone to damage to the inner vessel wall and not easy to use in vessels of smaller diameter.
In response to the above problems, foreign Zhongwei Jiang et al propose stirring and breaking thrombus and accelerating the dissolution rate of thrombus by driving mechanical vibration by a piezoelectric element. The specific method is to inject a small amount of thrombolytic agent into the thrombus blocking position, and stir and crush the thrombus through the vibration generated by the designed miniature vibration actuator, so as to fully combine with the thrombolytic agent, thereby achieving the purpose of rapid thrombolysis. On the basis of this point, yang Jingjing et al devised a miniature vibration actuator suitable for use in cerebral vessels.
The physical characteristics of thrombus are diversified due to different formation sites, formation reasons and formation time, and the required thrombolytic control parameters and thrombolytic drug dosage are different. In order to accelerate the thrombus removing time and better make a control strategy of the vibration actuator, it is important to obtain the thrombolysis effect of the actuator in the blood vessel in real time. The traditional thrombolysis effect prediction method has the technical problems of low reliability and low accuracy.
Disclosure of Invention
In view of the above, the application provides a method and a system for predicting the thrombolysis effect of the piezoelectric impedance based on the regression of a support vector machine, so as to solve the technical problems of low reliability and low accuracy in the prior art.
In a first aspect, an embodiment of the present invention provides a method for predicting a thrombolysis effect of a piezoelectric impedance based on regression of a support vector machine, including: testing the piezoresistance data corresponding to thrombolytic solutions with different concentrations by using a thrombus vibration actuator; carrying out standardized processing on the piezoresistor resistance data to obtain a training set and a testing set; the training set and the testing set are used as characteristic parameters and are input into a support vector machine regression model optimized by a gray wolf algorithm for training and testing, and a trained support vector machine regression model is obtained; predicting the blood concentration of the target to be detected based on the trained support vector machine regression model and the target piezoresistance data; the target piezoelectric impedance data are blood pressure resistance data acquired by the thrombus vibration actuator on the target to be detected; and predicting the thrombolysis effect of the target to be detected based on the blood concentration.
Further, the method for testing the piezoresistance data corresponding to thrombolytic solutions with different concentrations by using the thrombus vibration actuator comprises the following steps: proportioning and simulating thrombolytic solutions with different concentrations, and inserting the thrombus vibration actuator; sinusoidal driving signals with preset resonance frequency are input through a function generator and a power amplifier of the thrombus vibration actuator, so that the thrombus vibration actuator vibrates and the thrombolytic solutions with different concentrations are fully stirred; and measuring real part piezoresistance data of the thrombolytic solutions with different concentrations after stirring by using the impedance analyzer of the thrombus vibration actuator, and obtaining piezoresistance data corresponding to the thrombolytic solutions with different concentrations.
Further, the normalized processing of the piezoresistive data includes: and carrying out standardization processing on the piezoresistor resistance data based on a mean variance normalization method.
Further, the training set and the testing set are used as characteristic parameters and are input into a support vector machine regression model optimized by a wolf algorithm for training and testing, and the method comprises the following steps: carrying out mathematical analysis on the piezoresistor resistance data to construct a support vector machine regression model; establishing a gray wolf optimization algorithm and searching an optimal value of the regression model parameters of the support vector machine, wherein the method comprises the following steps: determining a target value function of a gray wolf algorithm; setting a search range of penalty factors and Gaussian kernel function parameters, the population size of the gray wolves and the maximum iteration times; randomly initializing the gray wolf population; classifying the initialized wolf population, sorting according to the calculated numerical value of the target value function of each wolf, and sequentially naming a, b, c, d from large to low according to the set population size; updating the positions of a, b and c wolves, calculating the positions of d wolves and the updated positions of a, b and c wolves, and updating the current position of d wolves according to the positions of three head wolves closest to d; and calculating and comparing the value of the optimal target value function of the new population, if the value is better than the value of the previous target value function value, updating the value of the target value function, adding one to the iteration number, and continuously performing position updating in a circulating way until the iteration number reaches the maximum iteration number.
Further, predicting the thrombolytic effect of the target to be tested based on the blood concentration includes: determining a target concentration interval in which the blood concentration is located; and determining the thrombolytic effect corresponding to the target concentration interval based on a corresponding relation table of the preset concentration interval and the thrombolytic effect.
In a second aspect, the embodiment of the invention further provides a piezoelectric impedance thrombolysis effect prediction system based on support vector machine regression, which comprises: the system comprises a testing module, a processing module, a training module, a first prediction module and a second prediction module; the testing module is used for testing the piezoresistance data corresponding to thrombolytic solutions with different concentrations by using the thrombus vibration actuator; the processing module is used for carrying out standardized processing on the piezoresistor resistance data to obtain a training set and a testing set; the training module is used for taking the training set and the testing set as characteristic parameters, inputting the characteristic parameters into a support vector machine regression model optimized by a wolf algorithm for training and testing, and obtaining a trained support vector machine regression model; the first prediction module is used for predicting the blood concentration of the target to be detected based on the trained support vector machine regression model and the target piezoresistance data; the target piezoelectric impedance data are blood pressure resistance data acquired by the thrombus vibration actuator on the target to be detected; the second prediction module is used for predicting the thrombolysis effect of the target to be detected based on the blood concentration.
Further, the test module is further configured to: proportioning and simulating thrombolytic solutions with different concentrations, and inserting the thrombus vibration actuator; sinusoidal driving signals with preset resonance frequency are input through a function generator and a power amplifier of the thrombus vibration actuator, so that the thrombus vibration actuator vibrates and the thrombolytic solutions with different concentrations are fully stirred; and measuring real part piezoresistance data of the thrombolytic solutions with different concentrations after stirring by using the impedance analyzer of the thrombus vibration actuator, and obtaining piezoresistance data corresponding to the thrombolytic solutions with different concentrations.
Further, the training module is further configured to: carrying out mathematical analysis on the piezoresistor resistance data to construct a support vector machine regression model; establishing a gray wolf optimization algorithm and searching an optimal value of the regression model parameters of the support vector machine, wherein the method comprises the following steps: determining a target value function of a gray wolf algorithm; setting a search range of penalty factors and Gaussian kernel function parameters, the population size of the gray wolves and the maximum iteration times; randomly initializing the gray wolf population; classifying the initialized wolf population, sorting according to the calculated numerical value of the target value function of each wolf, and sequentially naming a, b, c, d from large to low according to the set population size; updating the positions of a, b and c wolves, calculating the positions of d wolves and the updated positions of a, b and c wolves, and updating the current position of d wolves according to the positions of three head wolves closest to d; and calculating and comparing the value of the optimal target value function of the new population, if the value is better than the value of the previous target value function value, updating the value of the target value function, adding one to the iteration number, and continuously performing position updating in a circulating way until the iteration number reaches the maximum iteration number.
Further, the second prediction module is further configured to: determining a target concentration interval in which the blood concentration is located; and determining the thrombolytic effect corresponding to the target concentration interval based on a corresponding relation table of the preset concentration interval and the thrombolytic effect.
In a third aspect, an embodiment of the present invention further provides an electronic device, including: the device comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes the method for predicting the thrombolysis effect of the piezoelectric impedance based on the regression of the support vector machine when executing the computer program.
The invention provides a support vector machine regression-based method and a support vector machine regression-based system for predicting the thrombolysis effect, wherein the thrombolysis effect of a thrombus vibration actuator is affected by the outside to present nonlinear change, and a support vector machine regression-constructed thrombolysis effect prediction model optimized by using a gray wolf algorithm is used for training the concentration obtained by experiments and corresponding piezoresistance data, so that the blood concentration of the actuator after thrombolysis can be accurately predicted through the feedback piezoresistance, and compared with the prediction precision of a traditional regression-type support vector machine, the prediction precision of the support vector machine is obviously improved, and the technical problems of low reliability and low accuracy in the prior art are alleviated.
Drawings
In order to more clearly illustrate the embodiments of the application or the technical solutions in the prior art, the drawings that are needed in the detailed description of the embodiments and the prior art will be briefly described below, it being obvious that the drawings in the following description are some embodiments of the application and that other drawings may be obtained from these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method for predicting a thrombolysis effect of a piezoelectric impedance based on regression of a support vector machine, which is provided by an embodiment of the invention;
FIG. 2 is a schematic view of a thrombus vibration actuator according to an embodiment of the present invention;
FIG. 3 is a graph showing the comparison between the predicted value and the actual value of the regression of the conventional support vector machine according to the embodiment of the present invention;
FIG. 4 is a graph showing a comparison of predicted and actual values of a thrombus-stirring effect based on a support vector machine regression optimized by a gray-wolf algorithm;
fig. 5 is a schematic diagram of a piezoelectric impedance thrombolysis effect prediction system based on support vector machine regression according to an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the present invention for achieving the intended purpose, the following detailed description will refer to the specific implementation, structure, characteristics and effects according to the present invention with reference to the accompanying drawings and preferred embodiments.
Embodiment one:
fig. 1 is a flowchart of a method for predicting a thrombolysis effect of a piezoelectric impedance based on regression of a support vector machine according to an embodiment of the present invention. As shown in fig. 1, the method specifically includes the following steps:
step S102, testing the piezoresistance data corresponding to thrombolytic solutions with different concentrations by using a thrombus vibration actuator.
And step S104, carrying out standardization processing on the piezoresistor resistance data to obtain a training set and a testing set.
Optionally, in the embodiment of the invention, the method based on mean variance normalization performs standardization processing on solutions with different concentrations and piezoresistance data.
And S106, taking the training set and the testing set as characteristic parameters, and inputting the characteristic parameters into a support vector machine regression model optimized by a wolf algorithm for training and testing to obtain a trained support vector machine regression model.
Step S108, based on the trained support vector machine regression model and the target piezoresistance data, predicting the blood concentration of the target to be detected; the target piezoresistance data are the piezoresistance data of blood collected by the thrombus vibration actuator to the target to be tested.
Step S110, predicting thrombolysis effect of the target to be tested based on blood concentration.
The invention provides a support vector machine regression-based piezoelectric impedance thrombolysis effect prediction method, which is characterized in that the thrombolysis effect of a thrombus vibration actuator is affected by the outside and shows nonlinear change, a support vector machine regression-constructed thrombolysis effect prediction model optimized by using a gray wolf algorithm is used for training the experimentally obtained concentration and corresponding piezoresistance data, the piezoresistance resistance of the actuator after thrombolysis is fed back can be accurately predicted, the prediction precision of the support vector machine is obviously improved compared with that of the conventional regression-type support vector machine, and the technical problems of low reliability and low accuracy in the prior art are alleviated.
Specifically, step S102 further includes the steps of:
Step S1021, proportioning and simulating thrombolytic solutions with different concentrations, and inserting a thrombus vibration actuator.
Alternatively, fig. 2 is a schematic view of a thrombus vibration actuator provided in accordance with an embodiment of the present invention. As shown in fig. 2, in the embodiment of the present invention, the thrombus vibration actuator includes an impedance analyzer, an upper computer, a power amplifier and a signal generator; wherein the impedance analyzer comprises a piezoelectric crystal.
Step S1022, inputting sinusoidal driving signals with preset resonance frequency through a function generator and a power amplifier of the thrombus vibration actuator, so that the thrombus vibration actuator vibrates and sufficiently stirs thrombolytic solutions with different concentrations;
Step S1023, measuring real part piezoresistance data of the thrombolytic solutions with different concentrations after stirring by an impedance analyzer of the thrombus vibration actuator, and obtaining piezoresistance data corresponding to the thrombolytic solutions with different concentrations.
For example, in the embodiment of the invention, a micro thrombus vibration actuator is designed, an experimental platform for simulating thrombolysis of the actuator is built, corresponding thrombolysis effect tests are carried out in an optimal frequency range obtained through experiments, real part piezoelectric impedance values fed back by piezoelectric elements under 60 groups of different blood concentrations are obtained, all data are subjected to standardization processing, 2/3 of experimental data are randomly selected as a training set, and 1/3 of data are used as a testing set.
Specifically, in the process of collecting 60 groups of experimental data, a solution with a proportioned concentration is adopted, the tip of an actuator is inserted into the solution, a 50V sinusoidal signal with the resonance frequency of 659.8Hz is input through a signal generator and an amplifier, and the vibration actuator vibrates and uniformly stirs the solution. After stirring uniformly, an impedance analyzer is adopted to input a 1V sweep frequency signal, the sweep frequency is set to 600Hz-730Hz, the step length is 1.3Hz, the acquisition point is 101 points, and the real part piezoelectric impedance fed back by the piezoelectric crystal is taken as output. In the experiment, the real part piezoelectric impedance value of the piezoelectric crystal under different blood concentrations can be recorded and output in real time through the impedance analyzer connected with the piezoelectric crystal and stored in an upper computer. From the 60 experimental data sets, 2/3 experimental data sets are randomly selected as training sets, and the remaining 1/3 experimental data sets are used as test sets. And meanwhile, aiming at the dimension problem among variables in the sample, carrying out mean variance normalization processing on training and testing data.
Specifically, step S106 further includes the steps of:
Step S1061, carrying out mathematical analysis on the piezoresistance data to construct a support vector machine regression model;
step S1062, establishing a gray wolf optimization algorithm and finding an optimal value of the support vector machine regression model parameters, including:
determining a target value function of a gray wolf algorithm;
Setting a search range of penalty factors and Gaussian kernel function parameters, the population size of the gray wolves and the maximum iteration times;
randomly initializing the gray wolf population;
Classifying the initialized wolf population, sorting according to the calculated numerical value of the target value function of each wolf, and sequentially naming a, b, c, d from large to low according to the set population size;
updating the positions of a, b and c wolves, calculating the positions of d wolves and the updated positions of a, b and c wolves, and updating the current position of d wolves according to the positions of three head wolves closest to d;
And calculating and comparing the value of the optimal target value function of the new population, if the value is better than the value of the previous target value function value, updating the value of the target value function, adding one to the iteration number, and continuously performing position updating in a circulating way until the iteration number reaches the maximum iteration number.
Specifically, in the embodiment of the invention, mathematical analysis is carried out on the blood concentration and the real part piezoimpedance value fed back by the piezoelectric wafer, the corresponding piezoresistance thrombolysis effect prediction is carried out by adopting support vector machine regression, and a Gaussian kernel function is selected as a kernel function of the support vector machine regression; first, assume that a sample set containing n samples is givenWhere x i is the ith eigenvector and R represents the real set.
Mapping the sample set from the low-dimensional space to the high-dimensional space through nonlinearity, wherein a support vector regression expression of the nonlinearity mapping is expressed as follows:
Wherein: k (x i,x)＝φ(xi)φ(xj) is a kernel function, b is an intercept, a i and Is the Lagrangian multiplier, where x i and x j are the ith and jth eigenvectors.
In order to support vector machine regression with high prediction accuracy, a gaussian kernel function is introduced as a kernel function of support vector machine regression, the kernel function being expressed as:
Compared with the traditional regression model, only when the predicted value f (x) is equal to the true value y, the prediction is judged to be correct, namely the loss is zero, and the support vector machine regression tolerates a loss deviation epsilon between the true value and the predicted value, and when the predicted value is within the deviation zone, the prediction is also considered to be correct, so that the SVR problem can be expressed as:
Where C is a penalty factor and l ε is an ε -insensitive loss function, which can be expressed as:
Introducing a relaxation variable ζ i and Then formula (3) can be written as:
Where ε (ε > 0) is the maximum error allowed, and the introduction of the kernel function and Lagrangian multiplier, equation (5) can be expressed as the following equation:
Wherein alpha i and Expressed as a lagrangian multiplier, when the lagrangian function takes a minimum, the high-dimensional mapped support vector machine expression can be expressed as:
where K (x, x i)＝φ(xi)φ(xj) is the kernel function and b is the intercept.
Aiming at the problem of inconsistent dimension in the data, preprocessing the data by using a mean variance normalization formula:
Where x is the value to be normalized, μ is the sample mean, and S is the standard deviation of the sample.
The influence of the punishment factor C and the kernel function parameter gamma on the prediction model is considered to be large, so that the punishment factor C and the kernel function parameter gamma are optimized by using a gray wolf optimization algorithm.
After the prediction model is established, the real part piezoelectric impedance signal of the piezoelectric crystal of the thrombus vibration actuator is used as an input sample, the blood concentration value is used as an output sample, and 2/3 of the 60 groups of experimental data are used as large sample training data for improving the accuracy of prediction.
Specifically, in the embodiment of the invention, the optimization process of establishing a gray wolf optimization algorithm and carrying out the regression model parameters of the support vector machine is as follows:
Optimizing the punishment coefficient C and the kernel function parameter gamma, wherein the flow is as follows: 1) Determining a target value function of a gray wolf algorithm; 2) Setting a search range ori_w_range_c, ori_w_range_gamma, the population size num of the gray wolves and the maximum iteration number total_num of a penalty factor C and a Gaussian kernel function parameter gamma; 3) Randomly initializing the gray wolf population; 4) Classifying initialized wolf populations, sorting according to the calculated fitness function value of each wolf, and sequentially named a, b, c and d from large to small according to the set population sizes; 5) Updating the positions of the wolves such as a, b and c, calculating the positions of d and the wolves such as a, b and c after updating, and updating the current position of d wolves according to the positions of three head wolves closest to d; 6) And calculating and comparing the optimal target value function value of the new population, if the optimal target value function value is better than the previous target value function value, updating the target value function value, and simultaneously, enabling t=t+1 to turn to 5) to continue to execute. If the value of the target value is worse than the previous value of the target value, adding 1 to the set counter, and when the maximum iteration number total_nums is reached or the count of the counter reaches 20, exiting the algorithm; 7) And optimizing parameters of the regression type support vector machine by using a gray wolf algorithm.
Optionally, the method provided by the embodiment of the invention further includes: determining the model prediction accuracy of an established support vector machine regression (GWO-SVR) prediction model optimized by a gray wolf algorithm by using a decision coefficient R 2 and a mean square error E, wherein the expression of the decision coefficient R 2 and the mean square error E is as follows:
Fig. 3 is a graph of a predicted value and a true value of a conventional support vector machine regression (SVR) according to an embodiment of the present invention, and fig. 4 is a graph of a predicted value and a true value of a thrombus stirring effect based on a support vector machine regression optimized by a wolf algorithm according to an embodiment of the present invention, wherein the abscissa of fig. 3 and the ordinate of fig. 4 are dimensionless data after standardized processing. As shown in FIGS. 3 and 4, the GWO-SVR prediction algorithm in the embodiments of the present invention has significantly improved accuracy over conventional SVR. The mean square error E of GWO-SVR is reduced from 0.05483498191998703 of SVR to 0.02810084622159569 by 48.75%, and the decision coefficient R 2 is increased from 94.52% to 97.19%.
Specifically, step S110 further includes the steps of:
Step S1101, determining a target concentration interval in which the blood concentration is located;
Step S1102, determining a thrombolytic effect corresponding to the target concentration interval based on a preset concentration interval and thrombolytic effect correspondence table.
In the embodiment of the invention, after accurately predicting the concentration of the thrombolytic solution, an evaluation standard of thrombolytic effect can be preliminarily formulated according to the prediction result, the concentration of the predicted solution is lower thrombolytic effect between 0% and 20%, the concentration is medium thrombolytic effect between 20% and 50%, and the concentration is higher thrombolytic effect between 50% and 70%, as shown in table 1.
TABLE 1 correspondence table of preset concentration interval and thrombolytic effect
In conclusion, the accuracy of predicting the thrombolysis degree based on GWO-SVR prediction algorithm is higher than that of the SVR prediction algorithm which is the regression of the traditional support vector machine, and the solution concentration can be accurately predicted through piezoelectric impedance to be used as an important basis for evaluating the thrombolysis effect of the actuator.
Embodiment two:
fig. 5 is a schematic diagram of a piezoelectric impedance thrombolysis effect prediction system based on support vector machine regression according to an embodiment of the present invention. As shown in fig. 5, the system includes: test module 10, processing module 20, training module 30, first prediction module 40, and second prediction module 50.
Specifically, the test module 10 is used for testing the piezoresistance data corresponding to thrombolytic solutions with different concentrations by using a thrombus vibration actuator.
And the processing module 20 is used for carrying out standardized processing on the piezoresistor resistance data to obtain a training set and a testing set.
Optionally, in an embodiment of the present invention, the piezoresistance data is normalized based on a mean variance normalization method.
The training module 30 is configured to input the training set and the testing set as feature parameters into a support vector machine regression model optimized by a wolf algorithm for training and testing, so as to obtain a trained support vector machine regression model.
A first prediction module 40, configured to predict a blood concentration of a target to be measured based on the trained support vector machine regression model and the target piezoresistance data; the target piezoresistance data are the piezoresistance data of blood collected by the thrombus vibration actuator to the target to be tested.
A second prediction module 50, configured to predict a thrombolysis effect of the target to be tested based on the blood concentration.
The invention provides a support vector machine regression-based piezoelectric impedance thrombolysis effect prediction system, which is characterized in that the thrombolysis effect of a thrombus vibration actuator is affected by the outside and shows nonlinear change, a support vector machine regression-constructed thrombolysis effect prediction model optimized by using a gray wolf algorithm is used for training the experimentally obtained concentration and corresponding piezoresistance data, the piezoresistance resistance of the actuator after thrombolysis is fed back can be accurately predicted, the prediction precision of the support vector machine is obviously improved compared with that of the conventional regression-type support vector machine, and the technical problems of low reliability and low accuracy in the prior art are alleviated.
Specifically, the test module 10 is further configured to:
Proportioning and simulating thrombolytic solutions with different concentrations, and inserting a thrombus vibration actuator;
Sinusoidal driving signals with preset resonance frequency are input through a function generator and a power amplifier of the thrombus vibration actuator, so that the thrombus vibration actuator vibrates and sufficiently stirs thrombolytic solutions with different concentrations;
and measuring real-part piezoresistance data of the thrombolytic solutions with different concentrations after stirring by an impedance analyzer of the thrombus vibration actuator to obtain piezoresistance data corresponding to the thrombolytic solutions with different concentrations.
Specifically, training module 30 is further configured to:
carrying out mathematical analysis on the piezoresistance data, and constructing a support vector machine regression model;
Establishing a gray wolf optimization algorithm and searching an optimal value of a regression model parameter of a support vector machine, wherein the method comprises the following steps:
determining a target value function of a gray wolf algorithm;
Setting a search range of penalty factors and Gaussian kernel function parameters, the population size of the gray wolves and the maximum iteration times;
randomly initializing the gray wolf population;
Classifying the initialized wolf population, sorting according to the calculated numerical value of the target value function of each wolf, and sequentially naming a, b, c, d from large to low according to the set population size;
updating the positions of a, b and c wolves, calculating the positions of d wolves and the updated positions of a, b and c wolves, and updating the current position of d wolves according to the positions of three head wolves closest to d;
And calculating and comparing the value of the optimal target value function of the new population, if the value is better than the value of the previous target value function value, updating the value of the target value function, adding one to the iteration number, and continuously performing position updating in a circulating way until the iteration number reaches the maximum iteration number.
Specifically, the second prediction module 50 is further configured to:
determining a target concentration interval in which the blood concentration is located;
And determining the thrombolytic effect corresponding to the target concentration interval based on a corresponding relation table of the preset concentration interval and the thrombolytic effect.
The embodiment of the invention also provides electronic equipment, which comprises: the method for predicting the thrombolysis effect of the piezoelectric impedance based on the regression of the support vector machine in the first embodiment is realized when the processor executes the computer program.
The present invention is not limited to the above embodiments, but is capable of modification and variation in detail, and other modifications and variations can be made by those skilled in the art without departing from the scope of the present invention.
Claims (6)
1. A piezoelectric impedance thrombolysis effect prediction method based on support vector machine regression is characterized by comprising the following steps:
testing the piezoresistance data corresponding to thrombolytic solutions with different concentrations by using a thrombus vibration actuator;
Carrying out standardized processing on the piezoresistor resistance data to obtain a training set and a testing set;
the training set and the testing set are used as characteristic parameters and are input into a support vector machine regression model optimized by a gray wolf algorithm for training and testing, and a trained support vector machine regression model is obtained;
Predicting the blood concentration of the target to be detected based on the trained support vector machine regression model and the target piezoresistance data; the target piezoelectric impedance data are blood pressure resistance data acquired by the thrombus vibration actuator on the target to be detected;
predicting the thrombolysis effect of the target to be detected based on the blood concentration;
The training set and the testing set are used as characteristic parameters and are input into a support vector machine regression model optimized by a wolf algorithm for training and testing, and the method comprises the following steps:
Carrying out mathematical analysis on the piezoresistor resistance data to construct a support vector machine regression model;
establishing a gray wolf optimization algorithm and searching an optimal value of the regression model parameters of the support vector machine, wherein the method comprises the following steps:
determining a target value function of a gray wolf algorithm;
Setting a search range of penalty factors and Gaussian kernel function parameters, the population size of the gray wolves and the maximum iteration times;
randomly initializing the gray wolf population;
Classifying the initialized wolf population, sorting according to the calculated numerical value of the target value function of each wolf, and sequentially naming a, b, c, d from large to low according to the set population size;
updating the positions of a, b and c wolves, calculating the positions of d wolves and the updated positions of a, b and c wolves, and updating the current position of d wolves according to the positions of three head wolves closest to d;
Calculating and comparing the value of the optimal target value function of the new population, if the value is better than the value of the previous target value function value, updating the value of the target value function, adding one to the iteration number, and continuously performing position updating in a circulating way until the iteration number reaches the maximum iteration number;
Predicting the thrombolysis effect of the target to be detected based on the blood concentration, comprising:
Determining a target concentration interval in which the blood concentration is located;
and determining the thrombolytic effect corresponding to the target concentration interval based on a corresponding relation table of the preset concentration interval and the thrombolytic effect.
2. The method of claim 1, wherein testing the piezoresistive resistance data corresponding to thrombolytic solutions of different concentrations using a thrombotic vibration actuator comprises:
proportioning and simulating thrombolytic solutions with different concentrations, and inserting the thrombus vibration actuator;
Sinusoidal driving signals with preset resonance frequency are input through a function generator and a power amplifier of the thrombus vibration actuator, so that the thrombus vibration actuator vibrates and the thrombolytic solutions with different concentrations are fully stirred;
and measuring real part piezoresistance data of the thrombolytic solutions with different concentrations after stirring by using the impedance analyzer of the thrombus vibration actuator, and obtaining piezoresistance data corresponding to the thrombolytic solutions with different concentrations.
3. The method of claim 1, wherein normalizing the piezoresistive data comprises: and carrying out standardization processing on the piezoresistor resistance data based on a mean variance normalization method.
4. A piezoelectric impedance thrombolysis effect prediction system based on support vector machine regression is characterized by comprising: the system comprises a testing module, a processing module, a training module, a first prediction module and a second prediction module; wherein,
The test module is used for testing the piezoresistance data corresponding to thrombolytic solutions with different concentrations by using the thrombus vibration actuator;
the processing module is used for carrying out standardized processing on the piezoresistor resistance data to obtain a training set and a testing set;
the training module is used for taking the training set and the testing set as characteristic parameters, inputting the characteristic parameters into a support vector machine regression model optimized by a wolf algorithm for training and testing, and obtaining a trained support vector machine regression model;
The first prediction module is used for predicting the blood concentration of the target to be detected based on the trained support vector machine regression model and the target piezoresistance data; the target piezoelectric impedance data are blood pressure resistance data acquired by the thrombus vibration actuator on the target to be detected;
the second prediction module is used for predicting the thrombolysis effect of the target to be detected based on the blood concentration;
The training module is further configured to:
Carrying out mathematical analysis on the piezoresistor resistance data to construct a support vector machine regression model;
establishing a gray wolf optimization algorithm and searching an optimal value of the regression model parameters of the support vector machine, wherein the method comprises the following steps:
determining a target value function of a gray wolf algorithm;
Setting a search range of penalty factors and Gaussian kernel function parameters, the population size of the gray wolves and the maximum iteration times;
randomly initializing the gray wolf population;
Classifying the initialized wolf population, sorting according to the calculated numerical value of the target value function of each wolf, and sequentially naming a, b, c, d from large to low according to the set population size;
updating the positions of a, b and c wolves, calculating the positions of d wolves and the updated positions of a, b and c wolves, and updating the current position of d wolves according to the positions of three head wolves closest to d;
Calculating and comparing the value of the optimal target value function of the new population, if the value is better than the value of the previous target value function value, updating the value of the target value function, adding one to the iteration number, and continuously performing position updating in a circulating way until the iteration number reaches the maximum iteration number;
the second prediction module is further configured to:
Determining a target concentration interval in which the blood concentration is located;
and determining the thrombolytic effect corresponding to the target concentration interval based on a corresponding relation table of the preset concentration interval and the thrombolytic effect.
5. The system of claim 4, wherein the test module is further configured to:
proportioning and simulating thrombolytic solutions with different concentrations, and inserting the thrombus vibration actuator;
Sinusoidal driving signals with preset resonance frequency are input through a function generator and a power amplifier of the thrombus vibration actuator, so that the thrombus vibration actuator vibrates and the thrombolytic solutions with different concentrations are fully stirred;
and measuring real part piezoresistance data of the thrombolytic solutions with different concentrations after stirring by using the impedance analyzer of the thrombus vibration actuator, and obtaining piezoresistance data corresponding to the thrombolytic solutions with different concentrations.
6. An electronic device, comprising: a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method for predicting a thrombolytic effect of a piezoelectrical impedance based on support vector machine regression as claimed in any one of claims 1-3 when executing the computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211678107.2A CN115969464B (en) | 2022-12-26 | 2022-12-26 | Method and system for predicting thrombolysis effect of piezoelectric impedance based on regression of support vector machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211678107.2A CN115969464B (en) | 2022-12-26 | 2022-12-26 | Method and system for predicting thrombolysis effect of piezoelectric impedance based on regression of support vector machine |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115969464A CN115969464A (en) | 2023-04-18 |
| CN115969464B true CN115969464B (en) | 2024-05-10 |
Family
ID=85969523
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211678107.2A Active CN115969464B (en) | 2022-12-26 | 2022-12-26 | Method and system for predicting thrombolysis effect of piezoelectric impedance based on regression of support vector machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115969464B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5014715A (en) * | 1988-11-22 | 1991-05-14 | Chapolini Robert J | Device for measuring the impedance to flow of a natural or prosthetic vessel in a living body |
| JPH0422370A (en) * | 1990-05-17 | 1992-01-27 | Olympus Optical Co Ltd | Thrombus dissolutive therapy device |
| CN110956610A (en) * | 2019-10-31 | 2020-04-03 | 广州市番禺区中心医院(广州市番禺区人民医院、广州市番禺区心血管疾病研究所) | Method and system for predicting thrombolytic curative effect of lower limb deep venous thrombosis based on sparse representation |
| CN111145901A (en) * | 2019-12-04 | 2020-05-12 | 深圳大学 | Deep venous thrombosis thrombolytic curative effect prediction method and system, storage medium and terminal |
| CN113331912A (en) * | 2021-07-20 | 2021-09-03 | 昆明理工大学 | Vibration actuator for removing cerebral thrombosis and thrombus removing equipment |
| WO2022133258A1 (en) * | 2020-12-18 | 2022-06-23 | The Johns Hopkins University | Real-time prediction of adverse outcomes using machine learning |
| CN115101162A (en) * | 2022-06-16 | 2022-09-23 | 温州大学 | Thrombolysis efficiency prediction method and system based on biological fluid mechanics |
| CN115276452A (en) * | 2022-08-29 | 2022-11-01 | 苏州谱洛医疗科技有限公司 | Thrombolysis system circuit, ultrasonic energy adjusting method and ultrasonic thrombolysis system |
| CN115440383A (en) * | 2022-09-30 | 2022-12-06 | 中国医学科学院北京协和医院 | System for predicting curative effect of PD-1/PD-L1 monoclonal antibody of advanced cancer patient |
-
2022
- 2022-12-26 CN CN202211678107.2A patent/CN115969464B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5014715A (en) * | 1988-11-22 | 1991-05-14 | Chapolini Robert J | Device for measuring the impedance to flow of a natural or prosthetic vessel in a living body |
| JPH0422370A (en) * | 1990-05-17 | 1992-01-27 | Olympus Optical Co Ltd | Thrombus dissolutive therapy device |
| CN110956610A (en) * | 2019-10-31 | 2020-04-03 | 广州市番禺区中心医院(广州市番禺区人民医院、广州市番禺区心血管疾病研究所) | Method and system for predicting thrombolytic curative effect of lower limb deep venous thrombosis based on sparse representation |
| CN111145901A (en) * | 2019-12-04 | 2020-05-12 | 深圳大学 | Deep venous thrombosis thrombolytic curative effect prediction method and system, storage medium and terminal |
| WO2022133258A1 (en) * | 2020-12-18 | 2022-06-23 | The Johns Hopkins University | Real-time prediction of adverse outcomes using machine learning |
| CN113331912A (en) * | 2021-07-20 | 2021-09-03 | 昆明理工大学 | Vibration actuator for removing cerebral thrombosis and thrombus removing equipment |
| CN115101162A (en) * | 2022-06-16 | 2022-09-23 | 温州大学 | Thrombolysis efficiency prediction method and system based on biological fluid mechanics |
| CN115276452A (en) * | 2022-08-29 | 2022-11-01 | 苏州谱洛医疗科技有限公司 | Thrombolysis system circuit, ultrasonic energy adjusting method and ultrasonic thrombolysis system |
| CN115440383A (en) * | 2022-09-30 | 2022-12-06 | 中国医学科学院北京协和医院 | System for predicting curative effect of PD-1/PD-L1 monoclonal antibody of advanced cancer patient |
Non-Patent Citations (3)
| Title |
|---|
| 平均血小板体积预测静脉溶栓治疗的急性缺血性卒中患者的转归;张持;何晓璐;田玉蓉;俞刘珍熊;张康睿;杨若蕊;吴君仓;国际脑血管病杂志;20211231;第29卷(第003期);全文 * |
| 探究实时剪切波弹性成像技术预测静脉血栓硬度与溶栓疗效的相关性;夏晴;孙顺吉;马桂凤;张嫚嫚;林秀梅;;中国医疗设备;20171125(第11期);全文 * |
| 神经介入溶栓术对缺血性脑血管病患者血管内皮功能及血流动力学的影响探讨;刘宇;中国现代药物应用;20221130;第16卷(第21期);39-42 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115969464A (en) | 2023-04-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR102506092B1 (en) | Tensor field mapping | |
| JP6729455B2 (en) | Analytical data analysis device and analytical data analysis method | |
| Chinthrajah et al. | Development of a tool predicting severity of allergic reaction during peanut challenge | |
| Forshed et al. | Peak alignment of NMR signals by means of a genetic algorithm | |
| Tuckwell et al. | Analytical and simulation results for stochastic Fitzhugh-Nagumo neurons and neural networks | |
| Brown et al. | On the use of artificial neural networks for the analysis of survival data | |
| US20020145425A1 (en) | Methods for spectral analysis and their applications: spectral replacement | |
| JP6883574B2 (en) | Field-invariant and quantitative magnetic resonance signature | |
| CN112669960B (en) | Method for constructing liver fibrosis prediction model based on machine learning method, prediction system, equipment and storage medium | |
| Hou et al. | Prediction of firmness and pH for “golden delicious” apple based on elasticity index from modal analysis | |
| CN115969464B (en) | Method and system for predicting thrombolysis effect of piezoelectric impedance based on regression of support vector machine | |
| Shen et al. | Particle swarm optimization-based identification of the elastic properties in resonant ultrasound spectroscopy | |
| CN108414346B (en) | A kind of intelligent true triaxial test system and test method that test parameters is adaptive | |
| Gong et al. | Wearable microwave medical sensing for stroke classification and localization: A space-division-based decision tree learning method | |
| US10418129B2 (en) | Method and system for drug virtual screening and construction of focused screening library | |
| Ouyang et al. | Application of sparse linear discriminant analysis for metabolomics data | |
| CN115101162A (en) | Thrombolysis efficiency prediction method and system based on biological fluid mechanics | |
| Vang et al. | A combined deep learning-gradient boosting machine framework for fluid intelligence prediction | |
| Chen et al. | Prediction of dMRI signals with neural architecture search | |
| Mengash et al. | Automated autism spectral disorder classification using optimal machine learning model | |
| Euceda et al. | Prediction of clinical endpoints in breast cancer using NMR metabolic profiles | |
| Psuj | Multiple parameters fusion of electromagnetic nondestructive inspection data for evaluation of fatigue damage in steel elements | |
| Guimaraes et al. | Combining few neural networks for effective secondary structure prediction | |
| Nugraha et al. | Over-sampling strategies with data cleaning for handling imbalanced problems for diabetes prediction | |
| Tymkiv | Analysis of the complexity of algorithms for finding the coefficients of the mathematical model of lowintensity electroretinosignal |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |