A Medical Image Segmentation Method and System for Chronic Obstructive Pulmonary Disease

Abstract

The present invention provides a medical image segmentation method and system for chronic obstructive pulmonary disease, belonging to the field of medical image processing, including obtaining a chronic obstructive pulmonary disease image and converting it into a grayscale image, performing a non-local means filtering operation on the grayscale image to obtain a filtered image; constructing a corresponding two-dimensional histogram based on the filtered image and the grayscale image; introducing an adaptive transfer strategy and a spiral motion operator into the artemisinin algorithm to obtain an improved artemisinin algorithm; using Renyi entropy as the objective and the improved artemisinin algorithm as the search method to perform threshold search to obtain the optimal segmentation threshold combination; segmenting the chronic obstructive pulmonary disease image according to the optimal segmentation threshold combination and outputting the segmentation result. It provides accurate technical support for the early diagnosis of chronic obstructive pulmonary disease.

Description

Technical Field

[0001] The present invention belongs to the field of medical image processing, and particularly relates to a medical image segmentation method and system for chronic obstructive pulmonary disease. Background Art

[0002] Chronic Obstructive Pulmonary Disease (COPD) is a chronic respiratory disease. Early and accurate diagnosis and assessment of COPD are of great significance for the treatment and management of patients. However, due to problems such as blurred tissue boundaries in lung CT images and complex lesion morphologies, traditional imaging diagnosis methods are prone to misdiagnosis and low efficiency. Therefore, how to efficiently and accurately segment the lesion area in medical images has become an important research direction for COPD diagnosis.

[0003] Multi-threshold image segmentation, as an effective medical image processing technology, can divide an image into different regions by setting multiple thresholds, so as to accurately extract the target lesion. Its principle is based on the pixel gray histogram of the image, and by selecting an appropriate threshold combination to maximize a certain criterion (such as entropy), the best segmentation effect of the image is achieved. However, the main problem faced in multi-threshold segmentation is how to select a suitable threshold combination. Especially in a high-dimensional and complex threshold search space, traditional methods often have problems such as insufficient segmentation accuracy or long time consumption, and it is difficult to meet the actual needs. And meta-heuristic algorithms are often used for multi-threshold segmentation of medical images due to their powerful performance in global optimization problems. Of course, meta-heuristic algorithms also have problems such as being easily trapped in local optima and slow convergence speed, resulting in low accuracy in selecting threshold combinations when performing threshold segmentation on medical images, leading to inaccurate medical image segmentation and limiting their performance in practical applications. Summary of the Invention

[0004] In order to solve the multi-threshold segmentation problem of chronic obstructive pulmonary disease images, the present invention provides a medical image segmentation method and system for chronic obstructive pulmonary disease.

[0005] In order to achieve the above object, the present invention provides the following technical solutions:

[0006] A medical image segmentation method for chronic obstructive pulmonary disease specifically includes the following steps:

[0007] Obtain a medical image of chronic obstructive pulmonary disease, convert the medical image of chronic obstructive pulmonary disease into a grayscale image, perform a non-local means filtering operation on the grayscale image to obtain a filtered image; based on the filtered image and the grayscale image, obtain a two-dimensional histogram;

[0008] The adaptive transfer strategy and the spiral motion operator are introduced into the artemisinin algorithm to obtain an improved artemisinin algorithm; based on the two-dimensional histogram, the Renyi entropy with the maximum total entropy is taken as the target, and the improved artemisinin algorithm is used as the search method to obtain the optimal segmentation threshold combination;

[0009] The medical images of chronic obstructive pulmonary disease are segmented according to the optimal segmentation threshold combination, and the segmentation results are output.

[0010] Preferably, the improved artemisinin algorithm updates the individual positions through a spiral path, and the position update rule is specifically:

[0011] ;

[0012] ;

[0013] ;

[0014] ;

[0015] ;

[0016] where is the position of the rd individual in the th dimension, is the sorted population; is the distance between the population and the sorted population; is a constant equal to 1, which is used to control the shape of the logarithmic spiral function; represents the angular change part in the spiral path, so that the individual moves along a spiral shape in different dimensions; represents rounding; is the current evaluation number; and are process parameters; is the maximum evaluation number; represents the population size;

[0017] After the positions of the population individuals are updated, the migration process of the individuals is dynamically adjusted through adaptive perturbation, and the update rule of the individual migration is specifically:

[0018] ;

[0019] ;

[0020] ;

[0021] where is a perturbation amplitude control parameter; is a random number within the range of [-1, 1], which is used to introduce randomness in the search process; is used to adjust the position of the current solution; the individuals in the population need to be updated through and ; () represents the random function, p and is the randomly permuted dimension.

[0022] Preferably, perform non-local means filtering operation on the grayscale image to obtain the filtered image, specifically:

[0023] ;

[0024] where is the grayscale value of the pixel in the medical image of chronic obstructive pulmonary disease after denoising by non-local means filtering; refers to the corresponding weights of the pixels and , is the grayscale value of the pixel I in the medical image of chronic obstructive pulmonary disease f .

[0025] Preferably, based on the filtered image and the grayscale image, obtain a two-dimensional histogram. Specifically, the point is generated by the grayscale value of the pixel in the grayscale image and the grayscale value of the pixel in the non-local means image . Let represent the number of times the point appears. According to the joint probability density given by the following formula, obtain the two-dimensional histogram:

[0026] .

[0027] Preferably, take the Renyi entropy with the maximum total entropy as the objective based on the two-dimensional histogram, and use the improved artemisinin algorithm as the search method to obtain the optimal segmentation threshold combination. Specifically:

[0028] Let the gray level of the two-dimensional histogram be , represents a non-zero natural number. The pixel gray level and the non-local means gray level constitute a binary tuple . The expression of the Renyi entropy is:

[0029] ;

[0030] ;

[0031] ;

[0032] …

[0033] ;

[0034] ;

[0035] ;

[0036] …

[0037] ;

[0038] where represent different entropy values for image classification, represent the probabilities of different classes;

[0039] By setting the binary pair to maximize the Renyi total entropy, the maximum Renyi entropy is obtained;

[0040] Improve the artemisinin algorithm into a search method with the Renyi entropy with the maximum total entropy as the target, and obtain the optimal segmentation threshold combination.

[0041] The present invention also provides a medical image segmentation system for chronic obstructive pulmonary disease, specifically including:

[0042] An image processing module, configured to obtain a medical image of chronic obstructive pulmonary disease, convert the medical image of chronic obstructive pulmonary disease into a grayscale image, perform a non-local mean filtering operation on the grayscale image to obtain a filtered image; based on the filtered image and the grayscale image, obtain a two-dimensional histogram;

[0043] A threshold acquisition module, configured to introduce an adaptive transfer strategy and a spiral motion operator into the artemisinin algorithm to obtain an improved artemisinin algorithm; use the Renyi entropy with the maximum total entropy obtained based on the two-dimensional histogram as the target, and use the improved artemisinin algorithm as a search method to obtain the optimal segmentation threshold combination;

[0044] An image segmentation module, configured to segment the medical image of chronic obstructive pulmonary disease according to the optimal segmentation threshold combination and output a segmentation result.

[0045] The present invention also provides a computer device, including a memory, a processor, and a computer program stored on the memory. The processor executes the computer program to implement the steps in the medical image segmentation method for chronic obstructive pulmonary disease as described above.

[0046] The present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is loaded by a processor, it can execute the steps in the medical image segmentation method for chronic obstructive pulmonary disease as described above.

[0047] The medical image segmentation method for chronic obstructive pulmonary disease provided by the present invention has the following beneficial effects:

[0048] In the present invention, the obtained chronic obstructive pulmonary disease image is converted into a grayscale image and subjected to non-local means filtering. Then, the non-local means filtered image and the grayscale image are used to form a two-dimensional histogram. An adaptive transfer strategy and a spiral motion operator are introduced into the artemisinin algorithm, which improves the diversity and global search performance of the algorithm. The migration process of individuals is adjusted through dynamic perturbation, and it flexibly switches between global exploration and local exploitation at different stages. Renyi entropy is selected as the objective, and the improved artemisinin algorithm is used as the search method to obtain the optimal segmentation threshold combination of the global optimal solution, improving the accuracy of threshold combination selection. According to the obtained optimal threshold combination, the chronic obstructive pulmonary disease image is segmented to achieve rapid and accurate segmentation of the lesion area. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] To more clearly illustrate the embodiments of the present invention and their design schemes, the accompanying drawings required for the present embodiments will be briefly introduced below. The accompanying drawings in the following description are only partial embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0050] Figure 1 It is a flowchart of the medical image segmentation method for chronic obstructive pulmonary disease according to the present invention.

[0051] Figure 2 It is a plan view of the two-dimensional histogram in the medical image segmentation method for chronic obstructive pulmonary disease in the embodiments of the present invention.

[0052] Figure 3 It is a processing flowchart of the improved artemisinin algorithm (ADSPAO) in the embodiments of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0053] To enable those skilled in the art to better understand the technical solution of the present invention and to implement it, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the protection scope of the present invention.

[0054] Embodiment

[0055] The present invention provides a method for medical image segmentation of chronic obstructive pulmonary disease, as Figure 1 shown, specifically including the following steps:

[0056] S1: Obtain medical images of chronic obstructive pulmonary disease and initialize the threshold segmentation level.

[0057] S2: Convert the medical images of chronic obstructive pulmonary disease into grayscale images and perform non-local means filtering operations based on the grayscale images.

[0058] The non-local means filtering method is an effective method for removing noise. Specifically, assume and are the grayscale values of pixels I and e and f in the image I The non-local means of the image

[0059] ;

[0060] ;

[0061] ;

[0062] ;

[0063] ;

[0064] wherein, is the grayscale value of pixel after non-local means denoising, refers to the corresponding weights of pixels and , represents the normalization coefficient, refers to the control parameter, and are the local means, refers to the pixel block with a scale of around pixel , refers to the pixel block with a scale of around pixel If the image Gray-scale image has a size of , then the non-local means image obtained by non-local means filtering also has a size of , and are the gray-scale values of pixels , in the image, corresponding respectively to the gray-scale values of pixels in the local neighborhoods of pixels e and f , that is is located within a e region of size surrounding pixel , is a pixel within the neighborhood of pixel e , Similarly

[0065] S3: Based on the non-local means filtered image and the gray-scale image, construct the corresponding two-dimensional histogram. As Figure 2 shown, through the non-local means image and the gray-scale image, the corresponding non-local means two-dimensional histogram can be generated. The point is generated by the gray-scale value of pixel in the gray-scale image and the gray-scale value of pixel in the non-local means image . Denote the number of occurrences of point by . Its joint probability density is given by the following formula to obtain the final two-dimensional histogram

[0066] .

[0067] S4: Let the gray-scale levels of the two-dimensional histogram all be . The pixel gray-scale level and the non-local means gray-scale level form a binary tuple , then the expression of the Renyi entropy is

[0068] ;

[0069] ;

[0070] ;

[0071] …

[0072] ;

[0073] ;

[0074] ;

[0075] …

[0076] ;

[0077] wherein represent different entropy values for image classification, represent the probabilities of different classes.

[0078] For the binary pair set to maximize the Renyi total entropy, and the obtained .

[0079] S5: Introduce the adaptive transfer strategy and the spiral motion operator into the artemisinin algorithm to obtain the improved artemisinin algorithm.

[0080] S6: Use the improved artemisinin algorithm as the search method, and the processing flow is as Figure 3 shown. Take the Renyi entropy as the objective and obtain the optimal threshold combination based on the two-dimensional histogram.

[0081] S6.1: Initialize the parameters, including the population size , the maximum number of evaluations , the dimension , etc.

[0082] S6.2: Initialize the population as shown in the following formula. Calculate the fitness values of all individuals, that is, the Renyi entropy, and find the optimal individual and the optimal fitness value ;

[0083] ;

[0084] wherein, , is the population size; , is the dimension size; and respectively represent the maximum and minimum boundaries in the th dimension; is used to generate a random number between 0 and 1.

[0085] S6.3: Enter the comprehensive elimination stage, simulating the process of high-dose drugs rapidly spreading and widely acting on various parts of the human body during the early treatment of malaria. The mathematical model of this stage is as follows:

[0086] ;

[0087] Among them, represents the position of the -th individual in the -th generation in the dimension; represents the current best position; As a probability coefficient, it will be dynamically adjusted as the optimization process progresses. In the early stage of optimization, a larger value is set to strengthen the global search. In the later stage, is gradually decreased to shift to local fine optimization. is the decay exponent related to the drug concentration, simulating the exponential decay process of the drug concentration over time. The decay model of the drug concentration is as follows:

[0088] ;

[0089] ;

[0090] Among them, represents the drug concentration at time ; is the initial drug concentration; is the decay rate constant.

[0091] S6.4: Enter the local clearance stage, simulating the process of low-dose drugs continuing to act on the remaining pathogens in the later stage of treatment. The update process in this stage is as follows:

[0092] ;

[0093] ;

[0094] Among them, are three randomly selected search agents; is a random coefficient between [0.1, 0.6]; is the normalized fitness value of the -th search agent, representing the probability that this search agent is selected, and further optimizing the solution quality through local fine search; represents the fitness value of the -th search agent represent the minimum and maximum fitness values respectively.

[0095] S6.5: Enter the post-consolidation stage, simulating the possible risk of disease recurrence after the end of treatment. The update rule for this process is as follows:

[0096] .

[0097] S6.6: Calculate the fitness values of the updated population individuals, and sort the population individuals according to the fitness values. Introduce the spiral migration operator to update the individual positions through a spiral path to achieve the purpose of balancing exploration and exploitation. The update rules are as follows:

[0098] ;

[0099] ;

[0100] ;

[0101] ;

[0102] ;

[0103] where, is the sorted population; is the distance between the population and the sorted population; is a constant equal to 1, used to control the shape of the logarithmic spiral function; represents the angular change part in the spiral path, enabling the individual to move along a spiral shape in different dimensions; represents rounding; is the current evaluation count; and are process parameters.

[0104] S6.7: Execute the adaptive transfer strategy to dynamically adjust the individual migration process by introducing adaptive perturbations. To improve the global search ability of the artemisinin algorithm, a dynamic perturbation operator is added so that the perturbation intensity can gradually decay with the iteration process, thus achieving a gradual transition between exploration and exploitation. Specifically, the update of an individual can be described by the following formula:

[0105] ;

[0106] ;

[0107] ;

[0108] where, is a perturbation amplitude control parameter; is a random number in the range of [-1, 1], used to introduce randomness in the search process; is used to adjust the position of the current solution in some cases; the population individual needs to be updated through , and , and pRandom permutation dimension. This random dimension selection increases the diversity of the algorithm, enabling individuals to update different dimensions in different iterations, thereby enhancing the search ability.

[0109] S6.8: Use the greedy selection method to compare the newly generated solution with the initial solution. If the newly generated solution is better than the original solution, update it. At the same time, compare the updated solution with the optimal solution. If it is better than the optimal solution, update the optimal individual and the optimal fitness value bestfitness . That is, assume the fitness value of was originally 10, and the newly calculated fitness value is 100. Since 100 is greater than 10, will be replaced with = .

[0110] S6.9: Check if is satisfied. If it is, continue to execute S6.3; otherwise, output the optimal position and the optimal fitness value.

[0111] Output the optimal position and the optimal fitness value as the optimal threshold combination and the maximum Renyi entropy.

[0112] S7: Segment the medical images of chronic obstructive pulmonary disease according to the optimal threshold combination and output the segmentation result.

[0113] The following table shows the use of the Wilcoxon signed-rank test for three image quality evaluation metrics, namely the Peak Signal to Noise Ratio (PSNR), the Structural Similarity Index Peak Signal to Noise Ratio, PSNR), the Structural Similarity Index Structural Similarity (SSIM) and Feature Similarity Index (FSIM) are used to evaluate the quality of the segmented images. These three evaluation metrics are respectively used to assess the differences between the segmented images and the original images, measure the structural integrity of the images, and reflect the feature similarity between the original images and the segmented images. In the table, Thresholds represents the threshold level, Mean represents the average value of the overall ranking, Rank represents the ranking level, "+" indicates that the performance of ADSPAO is better than that of the comparison algorithm, "=" indicates that the performance of ADSPAO is the same as that of the comparison algorithm, and "-" indicates that the performance of ADSPAO is inferior to that of the comparison algorithm. According to the results of the Wilcoxon signed-rank test, it can be seen that at all threshold levels, the images segmented by the threshold set obtained using ADSPAO show superiority from multiple perspectives, with an overall average ranking of first, and have a certain degree of stability and robustness, and can effectively maintain the details and visual quality of the segmented images. CGPSO (Particle Swarm Optimization combining chaos and Gaussian local search procedure) is the particle swarm optimization combined with chaos and Gaussian local search process; MGSMA (Enhanced Slime Mould Algorithm based on a new movement mechanism and Gaussian kernel probability strategy) is the enhanced slime mould algorithm based on a new movement strategy and Gaussian kernel probability strategy; MDE (Modified Differential Evolution) is the improved differential evolution algorithm; IGWO (Improved Grey Wolf Optimizer) is the improved grey wolf optimization algorithm; GLSMA (Improved Slime Mould Algorithm based on Gaussian mutation and Lévy flights) is the improved slime mould algorithm based on Gaussian mutation and Lévy flights; WOA (Whale Optimization Algorithm) is the whale optimization algorithm.

[0114] Table 1 Comparison Results of PSNR

[0115]

[0116] Table 2 Comparison Results of SSIM

[0117]

[0118] Table 3 Comparison Results of FSIM

[0119]

[0120] The present invention also provides a medical image segmentation system for chronic obstructive pulmonary disease, specifically including:

[0121] An image processing module, configured to obtain a medical image of chronic obstructive pulmonary disease, convert the medical image of chronic obstructive pulmonary disease into a grayscale image, perform a non-local means filtering operation on the grayscale image to obtain a filtered image; and obtain a two-dimensional histogram based on the filtered image and the grayscale image.

[0122] A threshold acquisition module, configured to introduce an adaptive transfer strategy and a spiral motion operator into the artemisinin algorithm to obtain an improved artemisinin algorithm; take the Renyi entropy with the maximum total entropy as the target based on the two-dimensional histogram, and use the improved artemisinin algorithm as a search method to obtain an optimal segmentation threshold combination.

[0123] An image segmentation module, configured to segment the medical image of chronic obstructive pulmonary disease according to the optimal segmentation threshold combination and output a segmentation result.

[0124] Each module in the above-mentioned medical image segmentation system for chronic obstructive pulmonary disease can be implemented in whole or in part by software, hardware, and their combination. The above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or stored in a memory in a computer device in the form of software, so as to facilitate the processor to call and execute the operations corresponding to the above-mentioned modules.

[0125] The present invention also provides a computer device, including a memory, a processor, and a computer program stored on the memory. The processor executes the computer program to implement the steps in an embodiment of a medical image segmentation method for chronic obstructive pulmonary disease. The specific implementation method can be referred to the method embodiment and will not be elaborated here.

[0126] Furthermore, the present invention also provides a non-temporary computer-readable storage medium containing instructions, and a computer program is stored on the storage medium. For example, a memory containing instructions, and the above instructions can be executed by a processor of a computer device to complete the above method. For example, the non-temporary computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc. When the computer program is executed by the processor, it can implement the steps in an embodiment of a medical image segmentation method for chronic obstructive pulmonary disease. The specific implementation method can be referred to the method embodiment and will not be elaborated here.

[0127] Those skilled in the art should understand that the embodiments of the present invention may provide a method, a system or a computer program product. Therefore, the present invention may take the form of an all-hardware embodiment, an all-software embodiment or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0128] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems) and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of flows and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one or more flows Figure 1 or multiple flows and / or blocks.

[0129] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the functions specified in Figure 1 one or more flows Figure 1 or multiple flows and / or blocks.

[0130] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one or more flows Figure 1 or multiple flows and / or blocks.

[0131] It should be pointed out that the specific implementation methods described above can enable those skilled in the art to understand the invention more comprehensively, but do not limit the invention in any way. Therefore, although the invention has been described in detail in this specification and embodiments, those skilled in the art should understand that the invention can still be modified or replaced by equivalents; and all technical solutions and improvements that do not deviate from the spirit and scope of the invention are included in the protection scope of the patent for the invention. Any figure mark in the claims should not be regarded as limiting the claims involved. Any simple change or equivalent replacement of the technical solution that can be obviously obtained by any technician familiar with the field within the technical scope disclosed in the present invention belongs to the protection scope of the present invention.

Claims

1. A medical image segmentation method for chronic obstructive pulmonary disease, characterized in that, It includes the following steps: Obtain a medical image of chronic obstructive pulmonary disease, convert the medical image of chronic obstructive pulmonary disease into a grayscale image, perform a non-local mean filtering operation on the grayscale image to obtain a filtered image; Based on the filtered image and the grayscale image, obtain a two-dimensional histogram; Introduce an adaptive transfer strategy and a spiral motion operator into the artemisinin algorithm to obtain an improved artemisinin algorithm; take the Renyi entropy with the maximum total entropy based on the two-dimensional histogram as the objective, and use the improved artemisinin algorithm as the search method to obtain an optimal segmentation threshold combination; wherein, the improved artemisinin algorithm updates the individual position through a spiral path, and the position update rule is specifically: ； ； ； ； ； Among them, is the position of the -th individual in the dimension, is the sorted population; is the distance between the population and the sorted population; is a constant equal to 1, used to control the shape of the logarithmic spiral function; represents the angular change part in the spiral path, enabling the individual to move along a spiral shape in different dimensions; represents rounding; is the current number of evaluations; and are process parameters; is the maximum number of evaluations; represents the population size, is the individual update judgment parameter; After the population individual position is updated, dynamically adjust the individual migration process through adaptive perturbation, and the update rule of individual migration is specifically: ； ； ； Among them, is a perturbation amplitude control parameter; is a random number within the range of [-1, 1], which is used to introduce randomness in the search process; is used to adjust the position of the current solution; for each individual in the population needs to be updated through ; , () represents a random function, p is the dimension of random permutation, represents the current best position; Segment the medical image of chronic obstructive pulmonary disease according to the optimal segmentation threshold combination, and output the segmentation result.

2. The medical image segmentation method for chronic obstructive pulmonary disease according to claim 1, wherein, Perform a non-local mean filtering operation on the grayscale image to obtain a filtered image, specifically: ； Among them, is the gray value of the pixel in the medical image of chronic obstructive pulmonary disease after denoising by non-local means filtering; refers to the corresponding weights of pixels and ; is the gray value of the pixel I in the medical image of chronic obstructive pulmonary disease f .

3. A medical image segmentation method for chronic obstructive pulmonary disease according to claim 2, characterized in that, Based on the filtered image and the grayscale image, a two-dimensional histogram is obtained. Specifically, the point is composed of the pixel in the grayscale image and the pixel in the non-local mean image generated by the grayscale values. Let represent the number of occurrences of the point . According to the joint probability density given by the following formula, a two-dimensional histogram is obtained: ； Among them, is the total number of pixels of the image, and represent the number of pixels in the horizontal direction and the number of pixels in the vertical direction respectively.

4. A medical image segmentation method for chronic obstructive pulmonary disease according to claim 1, wherein Take the Renyi entropy with the maximum total entropy based on the two-dimensional histogram as the objective, and use the improved artemisinin algorithm as the search method to obtain an optimal segmentation threshold combination, specifically: Let the gray level of the two-dimensional histogram be , represent non-zero natural numbers, the pixel gray level and the non-local mean gray level form a binary tuple , and the expression of the Renyi entropy is as follows: ； ； ； … ； ； ； … ； wherein represent different entropy values for image classification, represent the probabilities of different classes, is the order of the Renyi entropy; By setting the pair to maximize the Renyi total entropy, the maximum Renyi entropy is obtained ; Use the improved artemisinin algorithm as the search method, take the Renyi entropy with the maximum total entropy as the objective, and obtain an optimal segmentation threshold combination.

5. A medical image segmentation system for chronic obstructive pulmonary disease, characterized in that, It includes: An image processing module, configured to obtain a medical image of chronic obstructive pulmonary disease, convert the medical image of chronic obstructive pulmonary disease into a grayscale image, perform a non-local mean filtering operation on the grayscale image to obtain a filtered image; Based on the filtered image and the grayscale image, obtain a two-dimensional histogram; A threshold acquisition module, configured to introduce an adaptive transfer strategy and a spiral motion operator into the artemisinin algorithm to obtain an improved artemisinin algorithm; take the Renyi entropy with the maximum total entropy based on the two-dimensional histogram as the objective, and use the improved artemisinin algorithm as the search method to obtain an optimal segmentation threshold combination; wherein, the improved artemisinin algorithm updates the individual position through a spiral path, and the position update rule is specifically: ； ； ； ； ； Among them, is the position of the -th individual in the dimension, is the sorted population; is the distance between the population and the sorted population; is a constant equal to 1, used to control the shape of the logarithmic spiral function; represents the angular change part in the spiral path, enabling the individual to move along a spiral shape in different dimensions; represents rounding; is the current evaluation count; and are process parameters; is the maximum evaluation count; represents the population size, is the individual update judgment parameter; After the population individual position is updated, dynamically adjust the individual migration process through adaptive perturbation, and the update rule of individual migration is specifically: ； ； ； Among them, is a perturbation amplitude control parameter; is a random number within the range of [-1, 1], which is used to introduce randomness in the search process; is used to adjust the position of the current solution; an individual in the population needs to be updated through and , () represents a random function, p is a randomly arranged dimension, represents the current best position; An image segmentation module, configured to segment the medical image of chronic obstructive pulmonary disease according to the optimal segmentation threshold combination, and output the segmentation result.

6. A computer device, comprising a memory, a processor, and a computer program stored on the memory, characterized in that, The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is loaded by the processor, it can execute the steps of the method according to any one of claims 1 to 4.

Images