CN114758023A - Stomach electrical impedance tomography method of self-adaptive genetic algorithm - Google Patents
Stomach electrical impedance tomography method of self-adaptive genetic algorithm Download PDFInfo
- Publication number
- CN114758023A CN114758023A CN202210329410.5A CN202210329410A CN114758023A CN 114758023 A CN114758023 A CN 114758023A CN 202210329410 A CN202210329410 A CN 202210329410A CN 114758023 A CN114758023 A CN 114758023A
- Authority
- CN
- China
- Prior art keywords
- stomach
- value
- genetic algorithm
- adaptive
- electrical impedance
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/003—Reconstruction from projections, e.g. tomography
- G06T11/005—Specific pre-processing for tomographic reconstruction, e.g. calibration, source positioning, rebinning, scatter correction, retrospective gating
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0033—Features or image-related aspects of imaging apparatus classified in A61B5/00, e.g. for MRI, optical tomography or impedance tomography apparatus; arrangements of imaging apparatus in a room
- A61B5/004—Features or image-related aspects of imaging apparatus classified in A61B5/00, e.g. for MRI, optical tomography or impedance tomography apparatus; arrangements of imaging apparatus in a room adapted for image acquisition of a particular organ or body part
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0059—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence
- A61B5/0073—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence by tomography, i.e. reconstruction of 3D images from 2D projections
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/05—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves
- A61B5/053—Measuring electrical impedance or conductance of a portion of the body
- A61B5/0536—Impedance imaging, e.g. by tomography
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Biophysics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Molecular Biology (AREA)
- Radiology & Medical Imaging (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Veterinary Medicine (AREA)
- Theoretical Computer Science (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Heart & Thoracic Surgery (AREA)
- Public Health (AREA)
- Pathology (AREA)
- Medical Informatics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Genetics & Genomics (AREA)
- Artificial Intelligence (AREA)
- Physiology (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention discloses a stomach electrical impedance tomography method of a self-adaptive genetic algorithm, which designs the coding and group parameters of the genetic algorithm by establishing a stomach two-dimensional model; designing a fitness function f of the genetic algorithm, selecting an operator, a self-adaptive crossover operator and a mutation operator for operation, and implementing an elite retention strategy; performing a decoding and calculation positive problem; and finally, reconstructing a stomach electrical impedance image according to an iteration result. The method has the advantages of high search efficiency and faster overall convergence speed; compared with the traditional genetic algorithm, the method improves the searching speed and the imaging quality.
Description
Technical Field
The invention relates to the field of electrical impedance tomography, in particular to a stomach electrical impedance tomography method of a self-adaptive genetic algorithm.
Background
Electrical Impedance Tomography (EIT) is a functional imaging technique, and the EIT includes two modes, namely dynamic EIT and static EIT, according to different imaging targets. Dynamic EIT is imaging the variation of impedance distribution under different conditions; static EIT is the imaging of the absolute value of the impedance distribution. The dynamic imaging can inhibit the common noise in the measuring process, and is relatively simple to realize; static imaging is more susceptible to reconstruction model errors and measurement noise and is more difficult to implement than dynamic EIT. The present document mainly studies static EIT image reconstruction. EIT image reconstruction is a nonlinear inverse problem of severe morbidity, especially in static EIT, the pathological characteristics are severe. The impedance image reconstruction performance of the electrical impedance tomography method of the adaptive Genetic Algorithm is superior to that of the traditional Genetic Algorithm (GA) under the condition of general noise resistance requirements.
Disclosure of Invention
The invention aims to provide a stomach electrical impedance tomography method of a self-adaptive genetic algorithm, which is used for solving the problems of more iteration times and low convergence rate of the traditional genetic algorithm; the ill-conditioned nonlinear inverse problem of electrical impedance tomography; traditional genetic algorithms are prone to the problem of locally optimal solutions.
In order to solve the problems, the invention provides a stomach electrical impedance tomography of a self-adaptive genetic algorithm, which is realized by the following technical scheme, and mainly comprises the following steps:
coding and group setting, setting a fitness function f, designing selection operator, self-adaptive crossover operator and mutation operator operations for the stomach electrical impedance tomography genetic algorithm, implementing an elite retention strategy, decoding and calculating a positive problem, and realizing stomach image reconstruction.
For the encoding operation, comprising:
the method comprises the steps of establishing a stomach two-dimensional model, expressing the electric conductivity value of each triangular unit in the stomach subdivision model as a gene, expressing the number of the triangular units as the length of the gene, and expressing the code of the gene by binary number.
The fitness function f is calculated by the following formula:
f is a fitness function, Vij(σ) to calculate the boundary voltage value, UijTo actually measure the boundary voltage value, | | | | non-conducting phosphor 2Is a two-norm expression.
Operating on the adaptive crossover operator, comprising:
cross probability PcControlling the probability of performing crossover operations on parents, PcThe larger the population, the more diverse the population will be, but the greater the likelihood that the original individual will be destroyed. For this purpose, adaptive crossover probability P is proposedcThe calculation formula of (2) is as follows:
Pc=k3,whenf'<fave (3)
f'、faveand fmaxRespectively a current individual fitness value, a population average fitness value and a population maximum fitness value, wherein a constant k1And k2Respectively ranges from 0.1 to 0.5]And [0.6 to 1]. According to the cross probability PcAnd (4) performing a cross operation, wherein the paired two individuals mutually exchange respective partial genes at a cross position to form two new individuals, and implementing an elite retention strategy in an algorithm iteration process.
For the decoding operation, comprising:
setting the conductivity value of the triangular unit in a stomach conductivity value range [ 0.5339-1 s/m ] by using a decoding formula, wherein the decoding calculation formula is as follows:
delta is the conductivity value after decoding, sigmamaxMaximum value in the stomach conductivity range, σminThe minimum value in the stomach conductivity range, x (i, n) is the ith row and nth column of the population, and L is the substring length of the gene. Solving the positive problem of the decoded conductivity value to obtain a boundary voltage value V ij(σ)。
The invention provides a stomach electrical impedance tomography method of a self-adaptive genetic algorithm, which is characterized by comprising the following steps of: the self-adaptive genetic algorithm is adopted for searching and optimizing, so that the problem of directly solving the nonlinear inverse problem of the ill condition can be avoided.
The invention provides a stomach electrical impedance tomography method of a self-adaptive genetic algorithm, which is characterized by comprising the following steps of: compared to a fixed cross probability PcUsing adaptive cross probability PcWith a greater probability P of crossing in the preceding periodcThe population tends to the optimal solution, and a smaller cross probability P can be adopted in the later periodcSo that better individuals can not damage better solutions through cross operation.
The invention provides a stomach electrical impedance tomography method of a self-adaptive genetic algorithm, which is characterized by comprising the following steps: the convergence rate of the traditional GA is improved by introducing the elite reservation strategy.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flow chart of an adaptive genetic algorithm;
FIG. 2 is a stomach two-dimensional field subdivision model diagram;
FIG. 3 is a stomach two-dimensional target field subdivision model diagram, wherein the value range of the conductivity value of a target area is [ 0.5339-1 s/m ], and the conductivity values of other units in a field are set to be 1 s/m;
FIG. 4 is a stomach EIT image using a conventional genetic algorithm;
fig. 5 is a gastric EIT image using an adaptive genetic algorithm.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to the accompanying drawings and examples.
The invention provides a stomach electrical impedance tomography method of a self-adaptive genetic algorithm, and the flow of the algorithm is shown in figure 1. Firstly, establishing a stomach two-dimensional model, then setting codes and group parameters, then calculating a fitness function f, carrying out operations of a selection operator, an adaptive crossover operator and a mutation operator of a stomach EIT genetic algorithm, implementing an elite retention strategy, carrying out decoding and positive calculation, and finally carrying out stomach EIT image reconstruction according to an iteration result. The invention improves the convergence rate and the global search capability of the traditional genetic algorithm.
S101, establishing a stomach two-dimensional model
The positive problem of EIT is to solve the potential distribution of the field domain knowing the conductivity σ distribution and the injected excitation current. In fact, the electromagnetic field boundary value problem is solved by adopting a numerical calculation method. The invention adopts a finite element method to solve the positive problem. Firstly, a two-dimensional model of the stomach field is established, then finite element discretization is carried out on the model, the stomach field is discretized into M triangular units, as shown in fig. 2, and the conductivity sigma (sigma) is equal to1,σ2,…,σM-1,σM) There are M elements, each element being the conductivity on a respective delta cell.
S102 encoding and group setting
Since genetic algorithms cannot directly deal with parameters of the problem space, they must be converted into genetic algorithm space, chromosomes composed of genes in a certain structure. And performing coding operation according to the number of triangle units of the stomach subdivision, wherein the electric conductivity value of the triangle unit is expressed as a gene, the number of the triangle units is expressed as a gene length, the coding of the gene is expressed by binary number, and the substring length L of the gene is set as an 8-bit binary string. Cross probability PcDesigned as adaptive cross probability, constant k1And k3Respectively ranges from 0.1 to 0.5]And [0.6~1]Probability of variation PmThe value range is [ 0.05-0.1% ]. The initial population consists of a set of randomly generated initial impedances (initial solutions).
S103, calculating a fitness function
And calculating a fitness function f of the randomly generated initial population, and obtaining the fitness value of each individual by calculating the fitness function f, so that the operator selection operation and the judgment of the iteration termination condition are facilitated. The setting of the objective function s affects the iteration termination condition of the genetic algorithm, and the calculation formula of the objective function s is set as follows:
wherein s is an objective function, Vij(σ) to calculate the boundary voltage value, UijIs the measured boundary voltage value. | | non-woven hair2Is a two-norm expression. In the reconstruction of the EIT image of the stomach, since the fitness function f and the objective function s are in a mapping relationship, the inverse of the objective function s is taken as the fitness function f, namely:
where f is the fitness function, Vij(σ) to calculate the boundary voltage value, UijThe measured boundary voltage value is obtained. | | non-woven hair2Is a two-norm expression.
S104 judging iteration termination condition
And after the fitness value of the current population is calculated, judging an iteration termination condition. The iteration termination conditions set in the method are two, the first condition is to judge the objective function value s of an individual in the population, and when the objective function value s of the individual in the population is smaller than a set value, the circulation is ended; the second condition is iteration times, and if the iteration times are larger than the iteration times, the loop is ended; when the two conditions are satisfied, the loop is ended, and a stomach EIT image is output.
S105 adaptive genetic operator operation
If the above mentioned devices areAnd if the two iteration termination conditions are not met, performing genetic selection operator, self-adaptive crossover operator and mutation operator operation. The selection operator operation herein uses a roulette method that enables selection of a new population based on a probability proportional to the fitness value f. To facilitate the roulette selection operation, the sum of fitness values fit of all individuals in the group is first calculated1Namely:
fit1=sum(f) (7)
where sum () is a summation formula and f is the fitness of all individuals in the population. Secondly, calculating the proportion fit of the fitness value of each individual in the population in the total fitness value2Namely:
therein, fitiIs the fitness value of the ith individual. The selection operator operation generates a value of 0,1 by random]Matrix of intervals, arranged in ascending order, and then having randomly generated values in turn sum with fit2Comparing the sizes, if the random number is less than fit2If so, the fit corresponding to the value is reserved2Of (a).
Adaptive crossover operator operation. The current optimal population is generated through the selection operator operation, and then the self-adaptive crossover operator operation is carried out. The method adopts a single-point crossing method, two individuals are randomly selected as crossed chromosomes in equal probability, a breakpoint is randomly selected in equal probability, and the right ends of the breakpoints of the chromosomes are exchanged, so that new offspring are generated. P cControlling the probability of performing crossover operations on parents, PcThe larger the population, the more diverse the population will be, but the greater the likelihood that the original individual will be destroyed. For this purpose, adaptive crossover probability P is usedcThe calculation formula of (2) is as follows:
Pc=k3,whenf'<fave (10)
wherein, f', faveAnd fmaxRespectively is a current individual fitness value, a population average fitness value and a population maximum fitness value. With respect to constant k1And k3Respectively ranges from 0.1 to 0.5]And [0.6 to 1]. The choice of these values is empirical and they may vary among different problems. The genetic strategy based on self-adaptation aims to use the relatively larger individual fitness value of the current two individuals as judgment, and if the individual fitness value is lower than the average fitness value of the current population, the larger cross probability P is takencIf the average fitness value of the individuals is higher than the average fitness value of the current group, a smaller cross probability P is takencNo crossover is performed for the best individual.
And (5) carrying out mutation operator operation. The current optimal population is generated through the operation of the self-adaptive crossover operator, and then the single-point mutation operation is carried out. The basic idea is to have a smaller mutation probability PmTo alter certain genes randomly selected as variations in the population. The specific operation is to randomly generate a value of 0,1 ]The matrix of intervals is arranged in ascending order, and then the randomly generated values are summed with the variation probability P in sequencemComparing the sizes, if the random number is less than the mutation probability PmIf so, carrying out mutation operation, wherein the mutation position is randomly selected, and rewriting the gene bit string of the mutation position as 0 into 1; similarly, the overwrite of 1 is 0.
S106 implementing the Elite Retention strategy
Recording the individuals with the optimal fitness value in the current population when the population fitness value f is calculated, and replacing the individuals with the optimal fitness value in the parent population with the individuals with the worst fitness value in the offspring population after the operations of a genetic selection operator, a self-adaptive crossover operator and a mutation operator.
S107, decoding operation and calculation positive problem
Since the genetic algorithm deals with genotype individuals, after the genetic operator operation is completed and the elite retention strategy is implemented, a decoding operation, i.e., the transformation of genotype individuals into phenotypic individuals, is required. Setting the conductivity value of the triangular unit in a stomach conductivity value range [ 0.5339-1 s/m ] by using a decoding formula, wherein the decoding calculation formula is as follows:
where δ is the decoded conductivity value, σmaxMaximum value in the stomach conductivity range, σ minThe minimum value in the stomach conductivity range, x (i, n) is the ith row and nth column of the population, and L is the substring length of the gene. Solving the positive problem of the decoded conductivity value to obtain a boundary voltage value Vij(σ)。
S108 stomach EIT image reconstruction
After the genetic operator is operated and the elite retention strategy is implemented, the positive problem is decoded and calculated to obtain the new generation of optimal population, the objective function is calculated in the new generation of optimal population, and then whether the iteration termination condition is met or not is judged, if the iteration termination condition is met, a stomach EIT image is output, as shown in fig. 4 and 5.
While the invention has been described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (5)
1. A stomach electrical impedance tomography method of a self-adaptive genetic algorithm is characterized in that a stomach two-dimensional universal model is established, a finite element method is adopted to calculate a positive problem, and the self-adaptive genetic algorithm is designed to solve an inverse problem, and mainly comprises the following steps:
Coding and group setting, setting a fitness function f, designing selection operator, self-adaptive crossover operator and mutation operator operations for the stomach electrical impedance tomography genetic algorithm, implementing an elite retention strategy, decoding and calculating a positive problem, and realizing stomach image reconstruction.
2. The method for stomach electrical impedance tomography with adaptive genetic algorithm as claimed in claim 1, wherein the encoding operation comprises:
the method comprises the steps of establishing a stomach two-dimensional model, expressing the electric conductivity value of each triangular unit in the stomach subdivision model as a gene, expressing the number of the triangular units as the length of the gene, and expressing the code of the gene by binary number.
4. The method of adaptive genetic algorithm stomach electrical impedance tomography as claimed in claim 1, wherein operating on the adaptive crossover operator comprises:
the cross probability Pc controls the probability of performing cross operation on a parent, the greater Pc, the more diversity of individuals in a group will increase, but the possibility that the original individual is damaged also increases, and therefore, a calculation formula of an adaptive cross probability Pc is provided as follows:
PC=k3,when f′<fave
Wherein, f', faveAnd fmaxRespectively the current individual fitness value and the population average fitnessThe stress value and the maximum population fitness value, wherein the value ranges of constants k1 and k3 are respectively [ 0.1-0.5 ]]And [0.6 to 1%]. According to the cross probability PCAnd (4) performing a cross operation, wherein the paired two individuals mutually exchange respective partial genes at a cross position to form two new individuals, and implementing an elite retention strategy in an algorithm iteration process.
5. The method of gastric electrical impedance tomography with adaptive genetic algorithm of claim 1, wherein the decoding operation comprises:
setting the conductivity value of the triangular unit in a stomach conductivity value range [ 0.5339-1 s/m ] by using a decoding formula, wherein the decoding calculation formula is as follows:
where δ is the decoded conductivity value, σmaxMaximum value in the range of gastric conductivity, σminThe minimum value in the stomach conductivity range, x (i, n) is the ith row and nth column of the population, and L is the substring length of the gene. Solving the positive problem of the decoded conductivity value to obtain a boundary voltage value Vij(σ)。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210329410.5A CN114758023A (en) | 2022-03-30 | 2022-03-30 | Stomach electrical impedance tomography method of self-adaptive genetic algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210329410.5A CN114758023A (en) | 2022-03-30 | 2022-03-30 | Stomach electrical impedance tomography method of self-adaptive genetic algorithm |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114758023A true CN114758023A (en) | 2022-07-15 |
Family
ID=82328878
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210329410.5A Pending CN114758023A (en) | 2022-03-30 | 2022-03-30 | Stomach electrical impedance tomography method of self-adaptive genetic algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114758023A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116578611A (en) * | 2023-05-16 | 2023-08-11 | 广州盛成妈妈网络科技股份有限公司 | Knowledge management method and system for inoculated knowledge |
-
2022
- 2022-03-30 CN CN202210329410.5A patent/CN114758023A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116578611A (en) * | 2023-05-16 | 2023-08-11 | 广州盛成妈妈网络科技股份有限公司 | Knowledge management method and system for inoculated knowledge |
CN116578611B (en) * | 2023-05-16 | 2023-11-03 | 广州盛成妈妈网络科技股份有限公司 | Knowledge management method and system for inoculated knowledge |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Huang et al. | Polygonal approximation using genetic algorithms | |
US10931027B2 (en) | Method for array elements arrangement of l-shaped array antenna based on inheritance of acquired character | |
CN109413710B (en) | Clustering method and device of wireless sensor network based on genetic algorithm optimization | |
CN107682117B (en) | Short code length LT code degree distribution design method based on improved chicken flock optimization algorithm | |
CN107122843A (en) | A kind of traveling salesman problem method for solving based on improved adaptive GA-IAGA | |
Yen et al. | Hierarchical genetic algorithm based neural network design | |
CN114758023A (en) | Stomach electrical impedance tomography method of self-adaptive genetic algorithm | |
CN109902808B (en) | Method for optimizing convolutional neural network based on floating point digital variation genetic algorithm | |
CN111242281A (en) | Weight optimization method for deep convolutional neural network | |
Malik | A study of genetic algorithm and crossover techniques | |
CN115248781A (en) | Combined test case generation method, device and equipment and readable storage medium | |
CN110807526A (en) | Quantum logic gate obtaining method and device for quantum state conversion | |
Chakrapani et al. | Genetic algorithm applied to fractal image compression | |
CN116756207A (en) | Network key node mining method based on discount strategy and improved discrete crow search algorithm | |
CN103116805B (en) | A kind of segmentation replacement method upgrading genetic groups | |
CN107171712B (en) | Method for selecting transmitting terminal transmitting antenna in large-scale multi-input multi-output system | |
CN114462771A (en) | Electricity utilization abnormity analysis method, device, equipment, medium and product | |
Malarz et al. | Dynamics in Eigen quasispecies model | |
CN114142467A (en) | Power distribution network photovoltaic maximum access capacity measuring and calculating method based on non-precise modeling power flow model | |
Geetha et al. | An observational analysis of genetic operators | |
CN109902007B (en) | Test case generation method based on point dyeing model | |
CN113141272A (en) | Network security situation analysis method based on iteration optimization RBF neural network | |
CN113344202A (en) | Novel deep multi-core learning network model training method, system and medium | |
CN111859807A (en) | Initial pressure optimizing method, device, equipment and storage medium for steam turbine | |
Vazquez-Rodriguez et al. | A genetic based technique for the determination of power system topological observability |
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 |