EIT imaging method and device based on improved sparrow search algorithm

Abstract

The invention relates to the technical field of image imaging, in particular to an EIT imaging method and device based on an improved sparrow search algorithm, and the method comprises the steps: obtaining initial data collected by EIT equipment; data processing is carried out on the initial data according to a target optimization algorithm to obtain target parameters, the target optimization algorithm is an improved sparrow search algorithm, and the target parameters comprise a target adaptive value and a corresponding first matrix; the first matrix is verified, the first matrix is determined as a target matrix, and the target matrix is used for EIT imaging calculation; and performing imaging calculation according to the initial data and the target matrix to generate a target image. According to the method, EIT imaging is carried out through the improved sparrow search algorithm, so that imaging has a better image reconstruction effect, and the contrast and definition of the image are improved.

Description

Technical Field

[0001] The present application relates to the field of image imaging technology, and in particular to an EIT imaging method and device based on an improved sparrow search algorithm. Background Art

[0002] At present, the main application of bioelectrical impedance tomography (EIT) equipment in the medical field is to generate ventilation images by collecting and measuring the ventilation status of patients. The Tikhonov regularization algorithm is usually used for image reconstruction. The basic principle of this method is to impose constraints on the solution in the iterative process by introducing a continuous quadratic function as a penalty function, thereby ensuring the stability of the solution process.

[0003] However, this method has a significant drawback: since the penalty function is continuous, its effect on the solution is similar to a smoothing filter, resulting in low contrast and blurred edges in the reconstructed image, especially the boundary of the target area cannot be clearly displayed. This low-contrast and blurred image quality is far from the high-resolution and high-contrast requirements of biomedical images and cannot meet the needs of clinical diagnosis. Summary of the invention

[0004] An object of an embodiment of the present invention is to provide an EIT imaging method and device based on an improved sparrow search algorithm, which is used to solve the technical problems of low contrast and blurred images caused by the Tikhonov regularization algorithm in the EIT imaging process.

[0005] In a first aspect, an embodiment of the present invention provides an EIT imaging method based on an improved sparrow search algorithm, the method comprising:

[0006] Obtain the initial data collected by the EIT equipment;

[0007] Processing the initial data according to a target optimization algorithm to obtain target parameters, wherein the target optimization algorithm is an improved sparrow search algorithm, and the target parameters include a target fitness value and a corresponding first matrix;

[0008] Verifying the first matrix to determine that the first matrix is ​​a target matrix, wherein the target matrix is ​​used for EIT imaging calculation;

[0009] An imaging calculation is performed according to the initial data and the target matrix to generate a target image.

[0010] In a second aspect, an embodiment of the present invention provides an EIT imaging device based on an improved sparrow search algorithm, the device comprising:

[0011] An acquisition unit, used for acquiring initial data collected by the EIT device;

[0012] A processing unit, configured to perform data processing on the initial data according to a target optimization algorithm to obtain target parameters, wherein the target optimization algorithm is an improved sparrow search algorithm, and the target parameters include a target fitness value and a corresponding first matrix;

[0013] A verification unit, used for verifying the first matrix and determining that the first matrix is ​​a target matrix, wherein the target matrix is ​​used for EIT imaging calculation;

[0014] A calculation unit is used to perform imaging calculation according to the initial data and the target matrix to generate a target image.

[0015] In a third aspect, an embodiment of the present invention provides a computer device, including:

[0016] at least one processor; and,

[0017] a memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to the first aspect.

[0019] In a fourth aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a processor, the processor executes the method as described in the first aspect.

[0020] In the scheme implemented by the EIT imaging method, device, equipment and storage medium based on the improved sparrow search algorithm, the method provides a basis for subsequent image reconstruction through the initial data collected from the EIT device. The initial data reflects the electrical impedance distribution of the target area and is the original material for imaging calculation; the initial data is processed by the improved sparrow search algorithm to obtain optimized target parameters. The improved sparrow search algorithm can effectively avoid local optimality during the processing process and improve the global search capability, thereby obtaining more accurate target fitness values ​​and the corresponding first matrix; the obtained first matrix is ​​strictly verified to ensure its accuracy and reliability. Once the matrix is ​​confirmed to be the target matrix, it can be used for EIT imaging calculation to ensure the accuracy of the imaging results; the improved sparrow search algorithm significantly improves the contrast and clarity of the image, and the generated target image is more in line with the high-precision requirements of medical imaging and can better assist clinical diagnosis and research. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0022] Figure 1 is a flow chart of an EIT imaging method based on an improved sparrow search algorithm in one embodiment of the present invention;

[0023] Figure 2 is a flow chart of an improved sparrow search algorithm in one embodiment of the present invention;

[0024] Figure 3 is a structural schematic diagram of an EIT imaging device based on an improved sparrow search algorithm in one embodiment of the present invention;

[0025] Figure 4 It is a schematic diagram of the structure of a computer device in one embodiment of the present invention. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical scheme and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0027] It should be noted that, if there is no conflict, the various features in the embodiments of the present invention can be combined with each other, all within the scope of protection of the present invention. In addition, although the functional modules are divided in the device schematic diagram and the logical order is shown in the flow chart, in some cases, the steps shown or described can be performed in a different order from the module division in the device or the flow chart. Furthermore, the words "first", "second", "third", etc. used in the present invention do not limit the data and execution order, but only distinguish the same items or similar items with basically the same functions and effects.

[0028] The technical solution of the present application can be applied to software upgrade scenarios of various EIR devices.

[0029] In view of this, the present application proposes an EIT imaging method based on an improved sparrow search algorithm to solve the above problems, which is described in detail below.

[0030] See also Figure 1 , Figure 1A flow chart of an EIT imaging method based on an improved sparrow search algorithm provided by an embodiment of the present invention, the method comprising the following steps:

[0031] S10. Acquire initial data collected by the EIT device.

[0032] Among them, electrical impedance tomography (EIT) is a type of non-invasive medical imaging. The basic principle is to apply a safe excitation current (voltage) signal to the surface of the target tissue to be measured, and simultaneously measure the voltage (current) signal on the surface of the target tissue. The impedance (or impedance change) image distribution in the detected tissue is obtained by using an image reconstruction algorithm based on the measured signal.

[0033] Therefore, the initial data may be raw voltage and current data obtained from the surface of the object being measured, as well as information such as electrode positions, etc. The initial data is usually a multi-dimensional matrix containing voltage measurements between electrode pairs.

[0034] S20. Process the initial data according to a target optimization algorithm to obtain target parameters. The target optimization algorithm is an improved sparrow search algorithm. The target parameters include a target fitness value and a corresponding first matrix.

[0035] Wherein, the target optimization algorithm is an improved sparrow search algorithm. The specific improvement steps are described between the following embodiments S201-S205. The sparrow search algorithm is a bionic optimization algorithm, which is inspired by the foraging behavior of sparrows and is used to solve complex optimization problems. The improved version may be optimized in terms of search efficiency, convergence speed, etc.

[0036] Among them, the target fitness value is an indicator to evaluate the quality of the solution. In EIT imaging, the first matrix (m 0 It is usually obtained based on prior knowledge or initial estimation. In the improved sparrow search algorithm, m 0 It can be used as the starting point of the optimization process and gradually improved through iterative updates of the algorithm.

[0037] Specifically, m 0 Used in the Tikhonov regularization formula, so that the Tikhonov regularization formula uses the target parameters for imaging.

[0038] Among them, the Tikhonov regularization formula is as follows:

[0039] m α,L =(H T H+αL T L) -1 (H Td+αL T Lm 0 ),

[0040] Among them, H is the Jacobi matrix, which represents the relationship between the initial data and the model parameters; d is the initial data vector; α is the regularization parameter, which controls the weight of the penalty function; L is the regularization matrix, which is used to introduce prior information; m α,L is the regularized solution; m 0 is the matrix corresponding to the initial estimates or prior values ​​of the model parameters.

[0041] Furthermore, the optimization goal is to minimize the norm function in the Tikhonov regularization formula, that is, to minimize the data residual and the penalty function at the same time.

[0042] It can be seen that in this embodiment, the target optimization algorithm can accurately obtain the target parameters including the target fitness value and the corresponding first matrix, effectively improve the image reconstruction effect of the Tikhonov regularization algorithm, enhance the contrast and clarity of the image, add the prior results of the optimization of non-human factors, reduce human intervention, and improve the objectivity and reliability of the imaging process.

[0043] S30: Verify the first matrix and determine that the first matrix is ​​a target matrix, where the target matrix is ​​used for EIT imaging calculation.

[0044] Verification is a key step to ensure that the first matrix meets certain quality standards and requirements. This may include checking the numerical stability of the matrix, whether it complies with physical constraints, whether it is within the expected parameter space, etc.

[0045] The verification process is as follows: Calculate the performance indicators of the first matrix, such as residuals and penalty functions, to ensure that these indicators are within an acceptable threshold range. Compare the current indicators with the previous results or expected values. If the current results are better, it can be considered that the matrix is ​​closer to the target matrix, and it is determined that the matrix can provide more accurate imaging results as the final matrix for EIT imaging calculation.

[0046] Among them, the target matrix is ​​a matrix that has been verified and confirmed, and it will be used as the basis for EIT imaging calculations. The accuracy of the target matrix directly affects the quality of the final imaging results. EIT imaging calculations are the process of reconstructing the electrical impedance distribution in the body using electrical impedance tomography technology. This process requires an accurate initial estimate (i.e., the target matrix) to improve the accuracy and stability of imaging.

[0047] It can be seen that the selected matrix in this embodiment can improve the accuracy and reliability of imaging, thereby achieving better diagnosis and analysis effects.

[0048] S40, performing imaging calculation according to the initial data and the target matrix to generate a target image.

[0049] Among them, imaging calculation is the process of reconstructing the internal electrical impedance distribution of the organism using the initial data and the target matrix through a specific algorithm (such as an inverse problem solving algorithm). This process usually involves the Tikhonov regularization formula.

[0050] The target image is the result of imaging calculation, which shows the electrical impedance distribution inside the organism in the form of an image. This image can be used for medical diagnosis, physiological monitoring and other applications.

[0051] It can be seen that the accuracy of the imaging results is ensured in this embodiment; the improved sparrow search algorithm significantly improves the contrast and clarity of the image, and the generated target image better meets the high-precision requirements of medical imaging.

[0052] The method provides a basis for subsequent image reconstruction through the initial data collected from the EIT device. The initial data reflects the electrical impedance distribution of the target area and is the original material for imaging calculations. The initial data is processed by the improved sparrow search algorithm to obtain optimized target parameters. The improved sparrow search algorithm can effectively avoid local optimality during the processing process and improve the global search capability, thereby obtaining more accurate target fitness values ​​and the corresponding first matrix. The obtained first matrix is ​​strictly verified to ensure its accuracy and reliability. Once the matrix is ​​confirmed to be the target matrix, it can be used for EIT imaging calculations to ensure the accuracy of the imaging results. The improved sparrow search algorithm significantly improves the contrast and clarity of the image, and the generated target image is more in line with the high-precision requirements of medical imaging and can better assist clinical diagnosis and research.

[0053] S201. In one embodiment, the processing the initial data according to the target optimization algorithm to obtain the target parameters includes: obtaining a second matrix, the second matrix is ​​an initial population, the initial population includes a plurality of sparrows, and the position of each sparrow corresponds to a row in the second matrix; calculating the fitness value of each sparrow in the second matrix according to the initial data; sorting the fitness value of each sparrow from large to small to obtain a first target sparrow with a maximum fitness value; obtaining a plurality of second target sparrows in the second matrix, the plurality of second target sparrows are all sparrows in the second matrix except the first target sparrow; randomly selecting an initial number of third target sparrows in the second matrix; performing data processing on the first target sparrow, the plurality of second target sparrows and the initial number of third target sparrows to obtain the target parameters, the target parameters including the target fitness value and the corresponding first matrix.

[0054] The process of obtaining the second matrix may be specifically referred to the step in S202.

[0055] Specifically, the second matrix represents the initial population, which is the starting point of the sparrow search algorithm, where each row corresponds to the position of a sparrow and represents a complete parameter vector, that is, each position represents a solution in the search space.

[0056] The initial population consists of multiple sparrows, and the position of each sparrow is determined by step S202, representing a point in the problem solution space.

[0057] Among them, the fitness function is usually based on the quality of EIT imaging, such as the contrast and clarity of the reconstructed image. The fitness evaluation will use the pre-processed EIT initial conductivity data to evaluate the performance of each parameter vector through imaging calculation.

[0058] Furthermore, in the process of calculating the fitness value of each sparrow in the second matrix according to the initial data, the initial data collected by the EIT device is used to define a fitness function, which reflects the degree of matching between the sparrow position (parameter vector) and the EIT data; the fitness value of each sparrow in the second matrix is ​​calculated according to the fitness function. The fitness function can be the objective function itself or a certain transformation form of the objective function.

[0059] Among them, the fitness values ​​of all sparrows are sorted from large to small. Then the sparrow with the highest fitness value (that is, the sparrow with the best performance) will be ranked first. The sparrow with the highest fitness value is selected from all sparrows as the discoverer, that is, the first target sparrow with the maximum fitness value, so the first target sparrow can be set as the discoverer. The first target sparrow means that the sparrow plays a leading role in the population, and its position is regarded as the current optimal or near-optimal solution.

[0060] Furthermore, the position information of the first target sparrow will affect the behavior and movement direction of other sparrows, helping the entire group to explore and develop the solution space more effectively in order to find the global optimal solution.

[0061] Among them, all sparrows except the first target sparrow in the second matrix are second target sparrows, so the first target sparrows can be set as joiners. Joiners are sparrows that follow the discoverer. The second target sparrows constitute the rest of the population.

[0062] An initial number of third target sparrows are randomly selected from the second matrix, and the third target sparrows can be set as guards. The third target sparrows provide additional diversity and help to jump out of the local optimum.

[0063] The initial number may be a randomly selected number and is not limited here.

[0064] Optionally, when calculating the fitness value of each sparrow, parallel computing can be used to improve efficiency.

[0065] Specifically, data processing is performed on the first target sparrow, the multiple second target sparrows and the initial number of third target sparrows to obtain target parameters, wherein the target parameters include a target fitness value and a corresponding first matrix. For the specific process, reference may be made to the description in subsequent S203-S205.

[0066] It can be seen that in this embodiment, the optimal solution is found by simulating the foraging behavior of sparrows, which is widely used in complex optimization problems. The calculation and update of the position and fitness value of each sparrow is the core of the algorithm, ensuring that the population can gradually approach the global optimal solution.

[0067] S202. In one embodiment, obtaining the second matrix includes: generating multiple chaotic sequences according to a preset chaotic mapping formula; filling each element of the initial matrix according to the multiple chaotic sequences to obtain a second matrix, wherein the second matrix is ​​an initial population, and the initial population includes multiple sparrows, and the position of each sparrow corresponds to a row in the second matrix.

[0068] The preset chaotic mapping formula may be a formula corresponding to a Cubic chaotic mapping, and the Cubic chaotic mapping has the characteristics of fast optimization speed and high precision.

[0069] Specifically, the preset chaotic mapping formula can be: Among them, ρ is the control function, which we take as 1 here; z n is the element in the initial matrix and also the nth Cubic chaotic mapping individual, that is, n represents the number of sparrow population.

[0070] Optionally, the control function can be tuned to the specific problem to obtain better chaotic properties.

[0071] When the population is initialized in S202, the matrix L generated by the chaotic mapping is used as the position or parameter of each sparrow in the sparrow algorithm.

[0072] Among them, a preset chaotic mapping formula is used to generate multiple chaotic sequences, which will be used to initialize the population. Chaotic sequences are highly random and unpredictable, which helps to increase the diversity of the population and avoid local optimal solutions.

[0073] Furthermore, the generated chaotic sequence is used to fill each element of the initial matrix, ensuring that the dimension of the matrix matches the parameter space in the EIT problem. For example, if the EIT problem has m parameters and the population size is N, the matrix should be m x N.

[0074] The dimension of the initial matrix (herein referred to as the “second matrix”) should match the parameter space in the EIT problem.

[0075] Assume that the EIT problem has 3 parameters (m=3) and the population size is 5 (N=5): generate 5 chaotic sequences, each with 3 values; fill these values ​​into a 3x 5 matrix to obtain a second matrix, each row of which represents the position of a sparrow, that is, a solution.

[0076] It can be seen that in this embodiment, chaotic mapping is used to initialize the population to increase diversity and avoid local optimal solutions, and has the characteristics of fast optimization speed and high precision.

[0077] S203. In one embodiment, the data processing of the first target sparrow, the plurality of second target sparrows and the initial number of third target sparrows to obtain the target parameters includes: performing position updating processing on the current position of the first target sparrow according to a first preset formula to obtain the first position of the first target sparrow; performing position updating processing on the current position of each second target sparrow in the plurality of second target sparrows according to a second preset formula to obtain the second position corresponding to each second target sparrow; performing quantity adjustment processing on the initial number of third target sparrows according to a third preset formula to obtain the target number of third target sparrows; performing position updating processing on the current position of each third target sparrow in the target number of third target sparrows according to a fourth preset formula to obtain the third position corresponding to each third target sparrow; combining the first position of the first target sparrow, the second position corresponding to each second target sparrow and the third position corresponding to each third target sparrow to obtain a third matrix; and obtaining the target parameters according to the third matrix, the target parameters including the target fitness value and the corresponding first matrix.

[0078] Among them, the first preset formula, the finder's location update formula is as follows:

[0079]

[0080] in,

[0081] Therefore, the position of the discoverer is updated according to the first preset formula, so that the "discoverer" is closer to the global optimal solution or explores a new potential solution area, and the updated "discoverer" position is output, which represents the new specific value of the sparrow in the solution space.

[0082] The first position is the new position of the first target sparrow after updating. The first position represents a new solution in the search space.

[0083] It is easy to fall into the local optimum during the iteration process, so the position of the joiner is updated after each iteration. When the sine strategy optimization algorithm is used, the global search capability is improved, and when the cosine strategy optimization algorithm is used, the local search capability is improved. The second preset formula is as follows:

[0084]

[0085] in,

[0086] · It means that in the t+1th iteration, the i-th sparrow is at the j-th dimension position.

[0087] · It means that in the tth iteration, the i-th sparrow is in the j-th dimension.

[0088] · Indicates the position of the sparrow with the best fitness value in the tth iteration.

[0089] x is a random number that conforms to the normal distribution and has a value range of [0,1].

[0090] S1 is a random number that conforms to the normal distribution and has a value range of [0,2].

[0091] S2 is a random number that conforms to the normal distribution and has a value range of [0,2π].

[0092] In the above formula, the position of the joiner is adjusted by the sine function to improve the global search capability or the local search capability. When x < 0.5, the value of the sine function is small, which is conducive to local search; when x ≥ 0.5, the value of the sine function is large, which is conducive to global search.

[0093] The second preset formula is applied to update the position of the joiner, and the second preset formula usually takes into account the new position of the discoverer to guide the joiner to move toward the discoverer or explore around it.

[0094] Furthermore, the process of updating the joiners is to ensure that other sparrows in the population can also move to the optimal solution area or explore new areas. In this way, the algorithm can maintain the diversity and exploration ability of the population while gradually converging to the optimal solution.

[0095] The second position is a new position of each second target sparrow obtained by applying the second preset formula. The second position represents multiple new solutions in the search space.

[0096] In each iteration cycle, a certain number of individuals are randomly selected as sentinels, and their number is adjusted according to the third preset formula. Then, according to the updated number of sentinels, the positions of the sentinels are updated according to the fourth preset formula to improve the global search capability of the algorithm and prevent falling into the local optimal solution.

[0097] Specifically, the third preset formula is as follows:

[0098]

[0099] in,

[0100] Among them, the fourth preset formula is as follows:

[0101] in,

[0102] Among them, the first position of the first target sparrow, the second position of each second target sparrow and the third position of each third target sparrow are combined to obtain a third matrix, which is a new population matrix and represents the overall solution space after the current iteration.

[0103] Among them, according to the third matrix, the target parameters are obtained. The specific implementation process of the target parameters including the target fitness value and the corresponding first matrix can refer to the process in S204, and will not be repeated here.

[0104] It can be seen that in this embodiment, the position of the first target sparrow is updated by the first preset formula, which can quickly guide the search to move to the potential optimal area and improve the search efficiency; the second preset formula is used to ensure that the solutions in the population are diverse, avoid premature convergence, and improve the global search capability; the third preset formula allows the number of third target sparrows to be dynamically adjusted according to the progress of the algorithm and the problem requirements, so that the population structure is more flexible and adapts to the search needs at different stages; the fourth preset formula updates the position of the third target sparrow, provides fine control over the search process, and helps to effectively utilize it while maintaining the exploration capability; by combining the positions of the first target sparrow, the second target sparrow and the third target sparrow to form a third matrix, the search results of different sparrows can be integrated to improve the quality of the final solution.

[0105] S204. In one embodiment, obtaining the target parameters based on the third matrix includes: comparing the third matrix with the second matrix to obtain a comparison value; if the comparison value is greater than a preset threshold, determining that the third matrix is ​​the target matrix; performing fitness calculation on the target matrix to obtain target parameters, the target parameters including the target fitness value and the corresponding first matrix.

[0106] The purpose of the comparison is to evaluate the degree of movement and change of the population during the search process. The specific comparison process is usually achieved by comparing the values ​​at each position in the new and old population matrices. If the new value at a certain position is better than the old value (for example, a better solution is found in the new population), then the new value will be retained; otherwise, if the old value is better, then the new value will be discarded and the old value will be retained. Through such a comparison and selection mechanism, the sparrow search algorithm can gradually approach the optimal solution to the problem.

[0107] The comparison value is a quantitative indicator used to measure the difference between the third matrix and the second matrix. This value can be based on a variety of metrics, such as Euclidean distance, similarity coefficient, etc.

[0108] The preset threshold is a pre-set critical value used to determine whether the comparison value is significant. If the comparison value exceeds this threshold, it means that the population position has changed sufficiently and may be close to or reach the optimal solution area. The setting of the preset threshold should be based on the characteristics of the problem, the number of iterations of the algorithm, and the expected solution quality.

[0109] If the comparison value is greater than a preset threshold, it is considered that the third matrix represents the effective movement of the population in the search space, and is therefore determined as the target matrix, that is, the optimal population position in the current search phase.

[0110] Among them, the fitness calculation is performed on the target matrix, that is, the quality of each sparrow position (solution) is evaluated. The fitness function is defined according to the specific problem and reflects the degree to which each solution is close to the optimal solution. The specific process can be referred to the description in S205.

[0111] It can be seen that the comparison and threshold judgment in this embodiment can be used as a monitoring mechanism in the algorithm iteration process, which is used to adjust the algorithm parameters in real time or terminate the iteration in advance to improve efficiency.

[0112] S205. In one embodiment, the performing fitness calculation on the target matrix to obtain target parameters includes: calculating the fitness value of each target sparrow in the target matrix; sorting the fitness value of each target sparrow from large to small to obtain a fourth target sparrow with a maximum fitness value; determining target parameters according to the fourth target sparrow, the target fitness value being the fitness value of the fourth target sparrow, and the first matrix being the position vector of the fourth target sparrow.

[0113] Among them, the sparrow with the highest fitness value in the sorted list is called the fourth target sparrow. This sparrow represents the optimal solution found in the current search process.

[0114] Among them, the target fitness value, i.e., the fitness value of the fourth target sparrow, represents the quality of the optimal solution; the first matrix, i.e., the position vector of the fourth target sparrow, represents the specific position of the sparrow in the search space when the optimal solution is reached.

[0115] Alternatively, the target parameters can be directly used for practical problems, such as parameter optimization in EIT imaging. The first matrix, as an estimate of the electrical impedance distribution, can be used to generate an imaging image.

[0116] Optionally, if the first matrix value is higher than the current m0, m0 is updated to the fitness value of the optimal solution.

[0117] For details, please refer to Figure 2 , Figure 2 This is a flowchart of an improved sparrow algorithm. Figure 2 In the process of initializing the population, it corresponds to S202, the fitness value calculation and sorting correspond to the steps in S201, the processes of updating the discoverer position, updating the joiner position, randomly selecting the sentinel and updating the position, and obtaining the updated position all correspond to the steps in S203, and the process of whether the updated position is better than the old position corresponds to the process in S205. Finally, the best fitness value and individual are output, which are the target parameters. The target fitness value is the fitness value of the fourth target sparrow.

[0118] It can be seen that in this embodiment, the optimal solution to the problem is found by using the optimized sparrow search algorithm, that is, by evaluating and selecting individuals with higher fitness values, the algorithm gradually approaches the optimal solution.

[0119] S206. In one embodiment, the verifying the first matrix to determine that the first matrix is ​​a target matrix includes: performing imaging calculation on the first matrix to obtain a first image and corresponding distribution data; performing residual calculation on the distribution data corresponding to the first image and the initial data to obtain a target residual; performing penalty function calculation on the first matrix to obtain a penalty function value; and verifying based on the target residual and the penalty function value to determine that the first matrix is ​​a target matrix.

[0120] The first matrix represents the optimal impedance distribution estimate obtained by an optimization algorithm (such as the sparrow search algorithm). Imaging calculation refers to using this impedance distribution estimate to generate a conductivity or resistivity image of the organism. This process usually involves numerical simulation and image reconstruction algorithms.

[0121] The first image is the result of the imaging calculation, showing the electrical impedance distribution inside the organism, and the corresponding distribution data is the electrical impedance value of each pixel or voxel in the image.

[0122] Among them, imaging calculation can adopt a variety of methods, such as finite element method, boundary element method, etc., depending on the design and requirements of the EIT system.

[0123] Among them, residual calculation refers to comparing the distribution data corresponding to the first image with the initial data and calculating the difference between the two. The initial data is usually the electrical data obtained by actual measurement. The residual reflects the degree of fit between the imaging result and the actual measured data. The residual calculation can be based on different metrics, such as mean square error, absolute error, etc.

[0124] The target residual is the result of residual calculation and is a quantitative indicator used to measure the accuracy of the imaging result. A smaller target residual means that the imaging result is closer to the actual measurement data.

[0125] Among them, the penalty function is a function used in constrained optimization problems, which aims to punish solutions that do not meet the constraints.

[0126] The penalty function value is the result of penalty function calculation, which reflects the degree to which the imaging result satisfies the constraint conditions. A lower penalty function value means that the imaging result is more in line with the constraint conditions.

[0127] The specific process of verifying according to the target residual and the penalty function value and determining that the first matrix is ​​the target matrix may refer to S207.

[0128] Among them, verification is performed based on the target residual and the penalty function value, that is, the quality of the imaging result is evaluated. If the verification result shows that the imaging result meets the preset standard (such as the target residual is small enough and the penalty function value is within an acceptable range), the first matrix is ​​confirmed as the target matrix, that is, it is considered to provide a reliable electrical impedance distribution estimation.

[0129] Furthermore, if the verification result does not meet the standard, it may be necessary to return to the optimization algorithm for further iterative optimization until a satisfactory result is obtained.

[0130] It can be seen that in this embodiment, the optimal solution or the solution that meets the conditions is determined through a series of calculation and verification steps.

[0131] S207. In one embodiment, the verification based on the target residual and the penalty function value to determine that the first matrix is ​​the target matrix includes: if the target residual is less than a first preset threshold value, and the penalty function value is less than a second preset threshold value, then judging and processing the target residual and the penalty function; obtaining historical data; if the target residual and the penalty function are both less than the historical data, then determining that the first matrix is ​​the target matrix.

[0132] The first preset threshold is a pre-set standard for judging whether the target residual is within an acceptable range. If the target residual is less than the first preset threshold, it means that the imaging result is consistent with the actual data.

[0133] The second preset threshold is another preset standard used to determine whether the penalty function value is within an acceptable range. If the penalty function value is less than the second preset threshold, it means that the imaging result satisfies the constraint condition well.

[0134] The first preset threshold and the second preset threshold may be set based on theoretical analysis, experimental verification or the experience of experts in the field to ensure the effectiveness and rationality of the thresholds.

[0135] The historical data refers to the records of target residuals and penalty function values ​​accumulated in previous imaging calculations. The historical data can be used to compare and evaluate the quality of the current imaging results.

[0136] The current target residual and penalty function values ​​are compared with the historical data. If the current values ​​are smaller than the historical data, it means that the current imaging result is better in quality than the previous attempt.

[0137] Wherein, when all the above conditions are met, the first matrix can be confirmed as the target matrix, which means that the first matrix provides a high-quality electrical impedance distribution estimation and can be used for subsequent imaging analysis and application.

[0138] For example, if the calculated residual and penalty function are less than a predetermined threshold, it is considered that a satisfactory solution has been found, the program ends, and the optimal fitness value and m0 are output. If the currently calculated residual and penalty function are smaller than the previous calculation result, or smaller than the predicted value, the current solution is considered better and the next iteration is performed. If the currently calculated residual and penalty function do not meet any of the above conditions, the current solution is considered not good enough and needs to be adjusted or the search process restarted. These two steps together constitute the convergence condition and stopping criterion of the algorithm, ensuring that the algorithm can find the global optimal solution within a reasonable time.

[0139] It can be seen that in this embodiment, the imaging results are judged and processed by comparing the target residual, the penalty function value and the historical data, so as to finally determine the first matrix as the target matrix, thereby ensuring the quality and reliability of the imaging results and providing a solid foundation for the application of EIT imaging technology.

[0140] S208. In one embodiment, performing imaging calculations based on the initial data and the target matrix to generate a target image includes: updating a preset regularization formula based on the target matrix to obtain a target regularization formula; performing imaging calculations based on the initial data and the target regularization formula to generate a target image.

[0141] Among them, the preset regularization formula is the Tikhonov regularization formula mentioned in S20 (as shown below), which will not be repeated here.

[0142] m α,L =(H T H+αL T L) -1 (H T d+αL T Lm 0 ).

[0143] The target matrix is ​​determined after optimization and verification, representing the optimal electrical impedance distribution estimate. According to this target matrix, the preset regularization formula needs to be updated to adapt to specific imaging conditions and requirements.

[0144] The imaging calculation is performed using the initial data (measured voltage or current data) and the target regularization formula. This process usually involves numerically solving the inverse problem to reconstruct the electrical impedance distribution inside the organism from the measured data.

[0145] Optionally, after the target image is generated, post-processing such as filtering, enhancement, segmentation, etc. may be required to improve the interpretability and practicality of the image.

[0146] It can be seen that in this embodiment, the target matrix can be used for EIT imaging calculation to ensure the accuracy of the imaging results; the improved sparrow search algorithm significantly improves the contrast and clarity of the image, and the generated target image is more in line with the high-precision requirements of medical imaging, which can better assist clinical diagnosis and research.

[0147] It should be noted that in each of the above-mentioned embodiments, there is not necessarily a certain order between the above-mentioned steps. A person of ordinary skill in the art can understand, based on the description of the embodiments of the present application, that in different embodiments, the above-mentioned steps may have different execution orders, that is, they may be executed in parallel, may be executed interchangeably, and so on.

[0148] As another aspect of the embodiment of the present application, the embodiment of the present application provides an EIT imaging device based on an improved sparrow search algorithm. The EIT imaging device based on the improved sparrow search algorithm can be a software module, which includes a number of instructions stored in a memory, and the processor can access the memory and call the instructions for execution to complete the EIT imaging method based on the improved sparrow search algorithm described in the above various embodiments.

[0149] See also Figure 3 , Figure 3 is a schematic diagram of the structure of an EIT imaging device based on an improved sparrow search algorithm provided in an embodiment of the present application. Figure 3 As shown, the EIT imaging device 300 based on the improved sparrow search algorithm includes:

[0150] An acquisition unit 301 is used for acquiring initial data collected by the EIT device;

[0151] A processing unit 302 is used to process the initial data according to a target optimization algorithm to obtain target parameters, wherein the target optimization algorithm is an improved sparrow search algorithm, and the target parameters include a target fitness value and a corresponding first matrix;

[0152] A verification unit 303, configured to verify the first matrix and determine that the first matrix is ​​a target matrix, wherein the target matrix is ​​used for EIT imaging calculation;

[0153] The calculation unit 304 is used to perform imaging calculation according to the initial data and the target matrix to generate a target image.

[0154] The method provides a basis for subsequent image reconstruction through the initial data collected from the EIT device. The initial data reflects the electrical impedance distribution of the target area and is the original material for imaging calculations. The initial data is processed by the improved sparrow search algorithm to obtain optimized target parameters. The improved sparrow search algorithm can effectively avoid local optimality during the processing process and improve the global search capability, thereby obtaining more accurate target fitness values ​​and the corresponding first matrix. The obtained first matrix is ​​strictly verified to ensure its accuracy and reliability. Once the matrix is ​​confirmed to be the target matrix, it can be used for EIT imaging calculations to ensure the accuracy of the imaging results. The improved sparrow search algorithm significantly improves the contrast and clarity of the image, and the generated target image is more in line with the high-precision requirements of medical imaging and can better assist clinical diagnosis and research.

[0155] In one embodiment, in the step of performing data processing on the initial data according to the target optimization algorithm to obtain target parameters, the processing unit 302 is further used to: obtain a second matrix, the second matrix is ​​an initial population, the initial population includes a plurality of sparrows, and the position of each sparrow corresponds to a row in the second matrix; calculate the fitness value of each sparrow in the second matrix according to the initial data; sort the fitness value of each sparrow from large to small to obtain a first target sparrow with a maximum fitness value; obtain a plurality of second target sparrows in the second matrix, the plurality of second target sparrows are all sparrows in the second matrix except the first target sparrow; randomly select an initial number of third target sparrows in the second matrix; perform data processing on the first target sparrow, the plurality of second target sparrows and the initial number of third target sparrows to obtain target parameters, the target parameters including the target fitness value and the corresponding first matrix.

[0156] In one embodiment, in the acquisition of the second matrix, the processing unit 302 is further used to: generate a plurality of chaotic sequences according to a preset chaotic mapping formula; fill each element of the initial matrix according to the plurality of chaotic sequences to obtain a second matrix, wherein the second matrix is ​​an initial population, and the initial population includes a plurality of sparrows, and the position of each sparrow corresponds to a row in the second matrix.

[0157] In one embodiment, in the step of performing data processing on the first target sparrow, the plurality of second target sparrows and the initial number of third target sparrows to obtain target parameters, the processing unit 302 is further used to: perform position update processing on the current position of the first target sparrow according to a first preset formula to obtain the first position of the first target sparrow; perform position update processing on the current position of each second target sparrow in the plurality of second target sparrows according to a second preset formula to obtain the second position corresponding to each second target sparrow; perform quantity adjustment processing on the initial number of third target sparrows according to a third preset formula to obtain the target number of third target sparrows; perform position update processing on the current position of each third target sparrow in the target number of third target sparrows according to a fourth preset formula to obtain the third position corresponding to each third target sparrow; obtain a third matrix by combining the first position of the first target sparrow, the second position corresponding to each second target sparrow and the third position corresponding to each third target sparrow; and obtain target parameters according to the third matrix, wherein the target parameters include a target fitness value and a corresponding first matrix.

[0158] In one embodiment, in obtaining the target parameters based on the third matrix, the processing unit 302 is further used to: compare the third matrix with the second matrix to obtain a comparison value; if the comparison value is greater than a preset threshold, determine that the third matrix is ​​the target matrix; perform fitness calculation on the target matrix to obtain target parameters, and the target parameters include a target fitness value and a corresponding first matrix.

[0159] In one embodiment, in performing fitness calculation on the target matrix to obtain target parameters, the processing unit 302 is further used to: calculate the fitness value of each target sparrow in the target matrix; sort the fitness value of each target sparrow from large to small to obtain a fourth target sparrow with a maximum fitness value; determine the target parameters according to the fourth target sparrow, the target fitness value is the fitness value of the fourth target sparrow, and the first matrix is ​​the position vector of the fourth target sparrow.

[0160] In one embodiment, during the verification of the first matrix, it is determined that the first matrix is ​​a target matrix, and the target matrix is ​​used in the EIT imaging calculation. The verification unit 303 is further used to: perform imaging calculation on the first matrix to obtain a first image and corresponding distribution data; perform residual calculation on the distribution data corresponding to the first image and the initial data to obtain a target residual; perform penalty function calculation on the first matrix to obtain a penalty function value; and perform verification based on the target residual and the penalty function value to determine that the first matrix is ​​the target matrix.

[0161] In one embodiment, in the step of verifying according to the target residual and the penalty function value to determine that the first matrix is ​​the target matrix, the verification unit 303 is further used to: if the target residual is less than a first preset threshold value, and the penalty function value is less than a second preset threshold value, perform judgment processing on the target residual and the penalty function; obtain historical data; if both the target residual and the penalty function are less than the historical data, determine that the first matrix is ​​the target matrix.

[0162] In one embodiment, in performing imaging calculations based on the initial data and the target matrix to generate a target image, the calculation unit 304 is further used to: update a preset regularization formula based on the target matrix to obtain a target regularization formula; and perform imaging calculations based on the initial data and the target regularization formula to generate a target image.

[0163] It should be noted that the above-mentioned EIT imaging device based on the improved sparrow search algorithm can execute the EIT imaging method based on the improved sparrow search algorithm provided in the embodiment of the present application, and has the corresponding functional modules and beneficial effects of the execution method. For technical details not fully described in the embodiment of the EIT imaging device based on the improved sparrow search algorithm, please refer to the EIT imaging method based on the improved sparrow search algorithm provided in the embodiment of the present application.

[0164] See also Figure 4 , Figure 4 4 is a schematic diagram of a computer device provided in an embodiment of the present application. The computer device includes one or more processors 41 and a memory 42. The memory 42 is connected to the one or more processors 41, for example, connected to the processor 41 via a bus.

[0165] The processor 41 is configured to support the computer device to perform the corresponding functions in the method in the above method embodiment. The processor 41 can be a central processing unit (CPU), a network processor (NP), a hardware chip or any combination thereof. The above hardware chip can be an application specific integrated circuit (ASIC), a programmable logic device (PLD) or a combination thereof. The above PLD can be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL) or any combination thereof.

[0166] The memory 42 is used to store program codes, etc. The memory 42 may include a volatile memory (VM), such as a random access memory (RAM); the memory 42 may also include a non-volatile memory (NVM), such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD) or a solid-state drive (SSD); the memory 42 may also include a combination of the above-mentioned types of memories.

[0167] The memory 42 can be used to store non-volatile software programs, non-volatile computer executable programs and modules, such as program instructions / modules corresponding to the EIT imaging method based on the improved sparrow search algorithm in the embodiment of the present application. The processor 41 executes various functional applications and data processing of the EIT imaging method based on the improved sparrow search algorithm and the EIT imaging device based on the improved sparrow search algorithm by running the non-volatile software programs, instructions and modules stored in the memory 42, that is, to realize the functions of the various modules or units of the EIT imaging method based on the improved sparrow search algorithm and the EIT imaging device based on the improved sparrow search algorithm provided in the above method embodiment.

[0168] The memory 42 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application required by at least one function. The data storage area may store data created according to the use of the EIT imaging device based on the improved sparrow search algorithm, etc. In some embodiments, the memory 42 may optionally include a memory 42 remotely arranged relative to the processor 41, and these remote memories 42 may be connected to the EIT imaging device based on the improved sparrow search algorithm via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0169] The one or more modules are stored in the memory 42. When executed by the one or more processors 41, the EIT imaging method based on the improved sparrow search algorithm in any of the above method embodiments is executed, for example, the method steps described in the above method embodiments are executed to realize the functions of the modules described in the above device embodiments.

[0170] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a computer, the computer executes the method described in the above embodiment.

[0171] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0172] The above disclosure is only the preferred embodiment of the present application, which certainly cannot be used to limit the scope of rights of the present application. Therefore, equivalent changes made according to the claims of the present application are still within the scope covered by the present application.

Claims

1. An EIT imaging method based on an improved sparrow search algorithm, characterized in that: include: Obtain the initial data collected by the EIT equipment; Processing the initial data according to a target optimization algorithm to obtain target parameters, wherein the target optimization algorithm is an improved sparrow search algorithm, and the target parameters include a target fitness value and a corresponding first matrix; Verifying the first matrix to determine that the first matrix is ​​a target matrix, wherein the target matrix is ​​used for EIT imaging calculation; An imaging calculation is performed according to the initial data and the target matrix to generate a target image.

2. The method according to claim 1, characterized in that The step of processing the initial data according to the target optimization algorithm to obtain target parameters includes: Acquire a second matrix, where the second matrix is ​​an initial population, the initial population includes a plurality of sparrows, and the position of each sparrow corresponds to a row in the second matrix; Calculate the fitness value of each sparrow in the second matrix according to the initial data; Sort the fitness values ​​of each sparrow from large to small to obtain the first target sparrow with the maximum fitness value; Acquire a plurality of second target sparrows in the second matrix, wherein the plurality of second target sparrows are all sparrows in the second matrix except the first target sparrow; Randomly selecting an initial number of third target sparrows in the second matrix; Data processing is performed on the first target sparrow, the plurality of second target sparrows and the initial number of third target sparrows to obtain target parameters, wherein the target parameters include a target fitness value and a corresponding first matrix.

3. The method according to claim 2, characterized in that The obtaining of the second matrix comprises: Generate multiple chaotic sequences according to the preset chaotic mapping formula; According to the multiple chaotic sequences, each element of the initial matrix is ​​filled to obtain a second matrix, where the second matrix is ​​an initial population. The initial population includes multiple sparrows, and the position of each sparrow corresponds to a row in the second matrix.

4. The method according to claim 2, characterized in that: The step of performing data processing on the first target sparrow, the plurality of second target sparrows and the initial number of third target sparrows to obtain target parameters includes: According to a first preset formula, a position update process is performed on the current position of the first target sparrow to obtain a first position of the first target sparrow; According to a second preset formula, respectively performing position update processing on the current position of each second target sparrow in the plurality of second target sparrows to obtain a second position corresponding to each second target sparrow; According to a third preset formula, the initial number of third target sparrows is subjected to quantity adjustment processing to obtain a target number of third target sparrows; According to the fourth preset formula, a position update process is performed on the current position of each third target sparrow in the target number of third target sparrows to obtain a third position corresponding to each third target sparrow; Combining the first position of the first target sparrow, the second position corresponding to each of the second target sparrows, and the third position corresponding to each of the third target sparrows, a third matrix is ​​obtained; According to the third matrix, target parameters are obtained, and the target parameters include a target fitness value and a corresponding first matrix.

5. The method according to claim 4, characterized in that The step of obtaining target parameters according to the third matrix includes: Compare the third matrix with the second matrix to obtain a comparison value; If the comparison value is greater than a preset threshold, determining the third matrix as a target matrix; The target matrix is ​​subjected to fitness calculation to obtain target parameters, wherein the target parameters include a target fitness value and a corresponding first matrix.

6. The method according to claim 5, characterized in that The step of performing fitness calculation on the target matrix to obtain target parameters includes: Calculating the fitness value of each target sparrow in the target matrix; Sort the fitness values ​​of each target sparrow from large to small to obtain a fourth target sparrow with the maximum fitness value; According to the fourth target sparrow, a target parameter is determined, the target fitness value is the fitness value of the fourth target sparrow, and the first matrix is ​​the position vector of the fourth target sparrow.

7. The method according to claim 1, characterized in that The verifying the first matrix to determine that the first matrix is ​​a target matrix includes: Performing imaging calculation on the first matrix to obtain a first image and corresponding distribution data; Performing residual calculation on the distribution data corresponding to the first image and the initial data to obtain a target residual; Performing penalty function calculation on the first matrix to obtain a penalty function value; Verification is performed based on the target residual and the penalty function value to determine that the first matrix is ​​a target matrix.

8. The method according to claim 7, characterized in that The verifying according to the target residual and the penalty function value to determine that the first matrix is ​​a target matrix includes: If the target residual is less than a first preset threshold value, and the penalty function value is less than a second preset threshold value, then judging and processing the target residual and the penalty function; Get historical data; If the target residual and the penalty function are both smaller than the historical data, the first matrix is ​​determined to be the target matrix.

9. The method according to claim 1, characterized in that: The performing imaging calculation according to the initial data and the target matrix to generate a target image includes: According to the target matrix, updating the preset regularization formula to obtain the target regularization formula; An imaging calculation is performed according to the initial data and the target regularization formula to generate a target image.

10. An EIT imaging device based on an improved sparrow search algorithm, the device comprising: An acquisition unit, used for acquiring initial data collected by the EIT device; A processing unit, configured to perform data processing on the initial data according to a target optimization algorithm to obtain target parameters, wherein the target optimization algorithm is an improved sparrow search algorithm, and the target parameters include a target fitness value and a corresponding first matrix; A verification unit, used for verifying the first matrix and determining that the first matrix is ​​a target matrix, wherein the target matrix is ​​used for EIT imaging calculation; A calculation unit is used to perform imaging calculation according to the initial data and the target matrix to generate a target image.

Images