A method for auxiliary diagnosis and treatment of temporomandibular joint based on medical imaging

Abstract

The present invention relates to the field of image processing technology, and in particular to a method for auxiliary diagnosis and treatment of temporomandibular joints based on medical images, comprising: clustering a temporomandibular joint grayscale image to obtain a target cluster; determining a connected domain in the target cluster and screening the target connected domain; performing superpixel segmentation on a region other than the target cluster to obtain a plurality of superpixel blocks; determining a judgment parameter according to the area of ​​the superpixel block; determining a position judgment parameter according to the centroid of the superpixel block and the target connected domain; obtaining the possibility that the superpixel block belongs to the articular disc region according to the area of ​​the superpixel block, the distance between the superpixel block and the centroid of the target connected domain, and the judgment parameter and position judgment parameter of the superpixel block; obtaining an updated grayscale value of a pixel point in each superpixel block according to the possibility and the pixel grayscale value; and enhancing the temporomandibular joint grayscale image according to the updated grayscale value. The present invention improves the clarity of the image and is helpful for formulating a treatment plan.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a method for auxiliary diagnosis and treatment of a temporomandibular joint based on medical imaging. Background Art

[0002] In order to assist in the diagnosis and treatment of TMJ-related diseases, a variety of auxiliary diagnosis and treatment methods based on medical imaging have been developed. Among them, common magnetic resonance imaging has many advantages in detecting TMJ, but it also has some problems and limitations. The most important problem with magnetic resonance imaging is that the resolution of soft tissue structures is not high enough, which makes it difficult to accurately locate these structures in medical images. The use of traditional histogram equalization image enhancement methods to uniformly enhance magnetic resonance imaging may reduce the contrast between soft tissue structures and surrounding tissues, causing segmentation difficulties, thereby increasing the difficulty of detecting and evaluating soft tissues in the TMJ. Summary of the invention

[0003] The present invention provides a medical imaging-based auxiliary diagnosis and treatment method for the temporomandibular joint to solve the existing problems.

[0004] The present invention provides a method for auxiliary diagnosis and treatment of the temporomandibular joint based on medical imaging, which adopts the following technical solutions:

[0005] An embodiment of the present invention provides a method for auxiliary diagnosis and treatment of the temporomandibular joint based on medical imaging, the method comprising the following steps:

[0006] Obtaining a grayscale image of the temporomandibular joint;

[0007] A clustering operation is performed on the grayscale image of the temporomandibular joint to obtain several clusters, and a target cluster is screened out from all the clusters; in the target cluster, all continuous adjacent pixels constitute a connected domain, and the target connected domain is screened out from all the connected domains according to the area of ​​the connected domain and the gradient values ​​of all pixels on the boundary;

[0008] Perform superpixel segmentation on the area except the target cluster in the grayscale image of the temporomandibular joint to obtain several superpixel blocks; divide each superpixel block into several reference sub-superpixel blocks and a neutron superpixel block; obtain the judgment parameter of each superpixel block according to the area size of the reference sub-superpixel block and the neutron superpixel block; obtain the position judgment parameter of each superpixel block according to the position of the centroid of the superpixel block and the target connected domain; obtain the possibility that each superpixel block belongs to the articular disc area according to the area of ​​the reference sub-superpixel block and the neutron superpixel block, the distance between the superpixel block and the centroid of the target connected domain, and the judgment parameter and position judgment parameter of each superpixel block;

[0009] According to the possibility that each superpixel block belongs to the articular disc area, the grayscale value of each pixel in the superpixel block, and the grayscale value of the pixel in the cluster where each pixel is located, the updated grayscale value of the pixel in each superpixel block is obtained;

[0010] In the temporomandibular joint grayscale image, the grayscale value of each pixel in each superpixel block is replaced with the updated grayscale value to obtain an enhanced temporomandibular joint grayscale image.

[0011] Preferably, the step of selecting a target cluster from all the clusters includes the following specific steps:

[0012] Calculate the mean grayscale value of all pixels in each cluster, and record the cluster with the largest mean grayscale value of all pixels as the target cluster.

[0013] Preferably, the method of selecting a target connected domain from all connected domains according to the area of ​​the connected domain and the gradient values ​​of all pixels on the boundary includes the following specific steps:

[0014] The connected domain of pixels that do not exist on the boundary of the temporomandibular joint grayscale image is recorded as the main connected domain;

[0015] Perform convex hull detection on all main connected domains to obtain the convex hull area, calculate the area ratio of each main connected domain in the convex hull area, and record the main connected domain whose area ratio in the convex hull area is greater than a preset threshold as the reference connected domain;

[0016] The target connected domain is screened out according to the gradient values ​​of all pixels on the boundary of the reference connected domain.

[0017] Preferably, the step of selecting the target connected domain according to the gradient values ​​of all pixels on the boundary of the reference connected domain comprises the following specific steps:

[0018] The Sobel operator is used to operate the grayscale image of the temporomandibular joint to obtain the gradient value of each pixel;

[0019] Calculate the mean of the gradient values ​​of all pixels on the boundary of the reference connected domain;

[0020] The reference connected domain with the largest mean of the gradient values ​​of all pixels on the boundary of the reference connected domain is recorded as the target connected domain.

[0021] Preferably, the step of dividing each super pixel block into a plurality of reference sub-super pixel blocks and a neutron super pixel block comprises the following specific steps:

[0022] The first The line connecting the two pixels with the longest distance in a superpixel block is recorded as the target straight line segment;

[0023] Divide the target straight line segment into three equal segments, and the corresponding two pixel points are recorded as segmentation pixel points;

[0024] The straight line that is perpendicular to the target straight line segment through the segmented pixel point is recorded as the segmentation line;

[0025] Use the dividing line to separate the The super pixel block is divided into three sub-super pixel blocks, and the sub-super pixel block between two segmentation lines is recorded as the neutron super pixel block;

[0026] Among the three sub-super-pixel blocks, the sub-super-pixel block that is not a neutron super-pixel block is recorded as a reference sub-super-pixel block.

[0027] Preferably, the step of obtaining the determination parameter of each superpixel block according to the area size of the reference sub-superpixel block and the neutron superpixel block comprises the following specific steps:

[0028] When the area of ​​any reference sub-superpixel block is greater than or equal to the area of ​​the neutron superpixel block, The determination parameter of the super pixel block is set to the first preset parameter T; when the area of ​​each reference sub-super pixel block in the reference sub-super pixel block is smaller than the area of ​​the neutron super pixel block, The determination parameter of each super pixel block is set to the second preset parameter E.

[0029] Preferably, the step of obtaining the position determination parameter of each superpixel block according to the position of the centroid of the superpixel block and the target connected domain comprises the following specific steps:

[0030] In the grayscale image of the temporomandibular joint, a rectangular coordinate system is constructed with the vertex in the lower left corner as the origin, the horizontal rightward as the horizontal axis, and the vertical upward as the vertical axis;

[0031] In the rectangular coordinate system, the value of the vertical coordinate of the pixel point is recorded as the height of the pixel point;

[0032] When the height of the centroid of the superpixel block is greater than the height of the centroid of the target connected domain, The position determination parameter of each super pixel block is set to a third preset parameter R;

[0033] When the height of the centroid of the superpixel block is less than or equal to the height of the centroid of the target connected domain, The position determination parameter of each super pixel block is set to the fourth preset parameter Q.

[0034] Preferably, the method of obtaining the possibility that each superpixel block belongs to the articular disc area according to the areas of the reference sub-superpixel block and the neutron superpixel block, the distance between the superpixel block and the centroid of the target connected domain, and the judgment parameter and position judgment parameter of each superpixel block includes the following specific calculation steps:

[0035]

[0036] Where: Indicates The probability that a superpixel block belongs to the articular disc area; Indicates The distance between the centroid of the superpixel patch and the centroid of the target connected domain; Represents the absolute value of the difference between the areas of two reference sub-superpixel blocks; Indicates The decision parameters of super pixel blocks, Indicates The position determination parameters of super pixel blocks.

[0037] Preferably, the method of obtaining the updated grayscale value of the pixels in each superpixel block according to the possibility that each superpixel block belongs to the articular disc area, the grayscale value of each pixel in the superpixel block, and the grayscale value of the pixels in the cluster where each pixel is located comprises the following specific steps:

[0038] The first In the superpixel block The maximum and minimum grayscale values ​​of all pixels in the cluster where the pixel is located are recorded as In the superpixel block The maximum grayscale value and the minimum grayscale value corresponding to the pixel points;

[0039] According to the possibility that each superpixel block belongs to the articular disc area, the grayscale value of each pixel in the superpixel block, and the maximum grayscale value and the minimum grayscale value corresponding to each pixel, the updated grayscale value of the pixel in each superpixel block is obtained.

[0040] Preferably, the method of obtaining the updated grayscale value of the pixels in each superpixel block according to the possibility that each superpixel block belongs to the articular disc area, the grayscale value of each pixel in the superpixel block, and the maximum grayscale value and the minimum grayscale value corresponding to each pixel, comprises the following specific steps:

[0041]

[0042] Where: Indicates In the superpixel block Updated grayscale value of pixels; Indicates In the superpixel block The minimum gray value of pixels; Indicates In the superpixel block The maximum gray value of pixels; Indicates The probability that a superpixel block belongs to the articular disc area; Indicates In the superpixel block The gray value of each pixel; represents the linear normalization function; Represents the ceiling function.

[0043] The beneficial effects of the technical solution of the present invention are: through clustering operations, different structures or tissues in the image can be separated, making subsequent processing more targeted; the identification of the target connected domain helps to locate the area of ​​the temporomandibular joint, thereby reducing the amount of calculation in subsequent processing and improving processing efficiency; determining the possibility of the articular disc area can help identify the key parts of the temporomandibular joint, thereby more accurately locating the problem and providing guidance for treatment; by updating the grayscale value of the pixel point, the contrast of the key structures in the image can be enhanced, the clarity and visualization of the image can be improved, and the generation of an enhanced image is helpful for formulating a treatment plan. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0045] Figure 1 The present invention is a flowchart of the steps of a method for auxiliary diagnosis and treatment of the temporomandibular joint based on medical imaging.

[0046] Figure 2 This embodiment provides a magnetic resonance medical image of the temporomandibular joint in the closed mouth position.

[0047] Figure 3 This is a grayscale image of the temporomandibular joint after grayscale adaptive equalization processing provided by this embodiment.

[0048] Figure 4 This is an image provided in this embodiment after clustering the grayscale image of the temporomandibular joint. DETAILED DESCRIPTION

[0049] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the specific implementation method, structure, characteristics and effects of a method for auxiliary diagnosis and treatment of the temporomandibular joint based on medical imaging proposed by the present invention are described in detail below in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0050] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0051] The following is a detailed description of a specific scheme of a medical imaging-based auxiliary diagnosis and treatment method for the temporomandibular joint provided by the present invention in conjunction with the accompanying drawings.

[0052] See also Figure 1 , which shows a flowchart of a method for auxiliary diagnosis and treatment of the temporomandibular joint based on medical imaging provided by an embodiment of the present invention, the method comprising the following steps:

[0053] Step S001: Acquire a grayscale image of the temporomandibular joint.

[0054] Use a 1.5T or higher magnetic resonance imaging device (MRI) to perform fast self-selected echo sequence T1-weighted imaging to obtain magnetic resonance medical images of the temporomandibular joint in the closed mouth position, such as Figure 2 The medical image is grayed out to obtain a grayscale image, and the grayscale image is further processed by grayscale adaptive equalization to obtain a grayscale image of the temporomandibular joint, as shown in Figure 3 Among them, grayscale adaptive equalization is a well-known technology, and the specific process will not be repeated in this embodiment.

[0055] It should be noted that the grayscale adaptive equalization process is to enhance the details and contrast of the image, making the anatomical structure clearer, easier to observe and diagnose. However, global enhancement may reduce the contrast between the soft tissue structure and the surrounding tissue, so it is necessary to further enhance the joint area in the image based on adaptive equalization.

[0056] Step S002: clustering the temporomandibular joint grayscale image to obtain several clusters, and selecting a target cluster from all the clusters; in the target cluster, all continuous adjacent pixel points constitute a connected domain, and the target connected domain is selected from all the connected domains according to the area of ​​the connected domain and the gradient values ​​of all pixel points on the boundary.

[0057] In routine MRI examinations of the temporomandibular joint, it is necessary to determine the position, morphology, displacement, and thickness of the articular disc to determine the corresponding symptoms. In the acquired images, the articular disc tissue and the surrounding soft tissue have similar low-grayscale appearances and are usually located in the center of the articular cavity of the temporomandibular joint and are not connected to other tissue structures. The condyle has a higher brightness and a higher grayscale appearance in the image. In addition, the condyle is a hemispherical structure, and its general appearance is a small circle or a nearly circular shape.

[0058] The Sobel operator is used to operate the temporomandibular joint grayscale image to obtain the gradient value of each pixel. The k-means clustering operation is performed on the temporomandibular joint grayscale image using the preset number of clusters to obtain several clusters. The clustered image is shown in Figure 4 The mean of the grayscale values ​​of all pixels in each cluster is calculated, and the cluster with the largest mean of the grayscale values ​​of all pixels is recorded as the target cluster. The k-means clustering operation is a well-known technology, and the specific process is not repeated in this embodiment.

[0059] In the target cluster, all consecutive adjacent pixels form a connected domain, and the connected domain of pixels that do not exist on the boundary of the temporomandibular joint grayscale image is recorded as the main connected domain. Convex hull detection is performed on all main connected domains to obtain the convex hull area, and the area ratio of each main connected domain in the convex hull area is calculated. The main connected domain whose area ratio in the convex hull area is greater than the preset threshold is recorded as the reference connected domain; the mean of the gradient values ​​of all pixels on the boundary of the reference connected domain is calculated.

[0060] The reference connected domain with the largest mean value of the gradient values ​​of all pixels on the boundary is recorded as the target connected domain. It should be noted that when there are multiple reference connected domains with the largest mean value of the gradient values ​​of all pixels on the boundary, one is selected from them.

[0061] It should be noted that the obtained target connected domain is the area where the condylar structure is located. In this example, the number of clusters preset is 3, and the preset threshold is 0.9. Other embodiments may use other values, which are not specifically limited in this embodiment.

[0062] Step S003: Perform superpixel segmentation on the area except the target cluster in the grayscale image of the temporomandibular joint to obtain several superpixel blocks; divide each superpixel block into several reference sub-superpixel blocks and one neutron superpixel block; obtain the judgment parameter of each superpixel block according to the area size of the reference sub-superpixel block and the neutron superpixel block; obtain the position judgment parameter of each superpixel block according to the position of the centroid of the superpixel block and the target connected domain; obtain the possibility that each superpixel block belongs to the articular disc area according to the area of ​​the reference sub-superpixel block and the neutron superpixel block, the distance between the superpixel block and the centroid of the target connected domain, and the judgment parameter and position judgment parameter of each superpixel block.

[0063] The main body of the articular disc is mainly composed of dense connective tissue, and appears as a dumbbell-shaped or bow-tie-shaped upper line biconcave black strip in the image, showing that the area at both ends is larger than that in the middle part, and is located above the condyle.

[0064] Perform superpixel segmentation on the area except the target cluster in the grayscale image of the temporomandibular joint to obtain several superpixel blocks. Take the superpixel block as an example, The line connecting the two pixels with the longest distance in a superpixel block is recorded as the target straight line segment; the two pixels corresponding to the target straight line segment being divided into three equal segments are recorded as segmentation pixels; the straight line passing through the segmentation pixel point and perpendicular to the target straight line segment is recorded as the segmentation line.

[0065] Use the dividing line to separate the The superpixel block is divided into three sub-superpixel blocks, and the sub-superpixel block between the two dividing lines is recorded as the neutron superpixel block; among the three sub-superpixel blocks, the sub-superpixel block that is not a neutron superpixel block is recorded as a reference sub-superpixel block. When the area of ​​the reference sub-superpixel block is larger than the neutron superpixel block, the possibility that the superpixel block belongs to the articular disc is greater.

[0066] In the grayscale image of the temporomandibular joint, a rectangular coordinate system is constructed with the vertex in the lower left corner as the origin, the horizontal rightward as the horizontal axis, and the vertical upward as the vertical axis. In the rectangular coordinate system, the horizontal coordinate of the pixel point is represented as its horizontal position in the image, the vertical coordinate is represented as the vertical position of the pixel point, and the value of the vertical coordinate of the pixel point is recorded as the height of the pixel point.

[0067] In this embodiment, when the area of ​​any reference sub-super pixel block is greater than or equal to the area of ​​the neutron super pixel block, The determination parameter of the super pixel block is set to the first preset parameter T; when the area of ​​each reference sub-super pixel block in the reference sub-super pixel block is smaller than the area of ​​the neutron super pixel block, The determination parameter of each super pixel block is set to the second preset parameter E. In this embodiment, the value of T is 0.1, and the value of E is 0.9. This is used as an example for description. Other implementations may be set to other values, which are not limited in this embodiment.

[0068] In this embodiment, when When the height of the centroid of the superpixel block is greater than the height of the centroid of the target connected domain, The position determination parameter of the super pixel block is set to the third preset parameter R; when the height of the centroid of the super pixel block is less than or equal to the height of the centroid of the target connected domain, the first The position determination parameter of each super pixel block is set to a fourth preset parameter Q. In this embodiment, the value of R is 0, and the value of Q is 1. This is used as an example for description. Other implementations may be set to other values, which are not limited in this embodiment.

[0069] This embodiment provides a calculation method for obtaining the possibility that a super pixel block belongs to the articular disc area:

[0070]

[0071] Where: Indicates The probability that a superpixel block belongs to the articular disc area; Indicates The distance between the centroid of the superpixel patch and the centroid of the target connected domain; Represents the absolute value of the difference between the areas of two reference sub-superpixel blocks; Indicates The decision parameters of super pixel blocks, Indicates The position determination parameters of super pixel blocks.

[0072] In the above formula, It means that the distance between the superpixel block and the target connected domain is used as the judgment basis. When the distance is closer, the possibility that the superpixel block belongs to the articular disc is higher. When the distance is farther, the possibility that the superpixel block belongs to the articular disc is lower. It indicates the possibility of judging the superpixel block belonging to the articular disc by the size of the two reference sub-superpixel blocks. The more similar the two reference sub-superpixel blocks are, the greater the possibility that the superpixel block belongs to the articular disc. In this embodiment, the denominator is added with 1 to ensure that the denominator is not zero.

[0073] At this point, the possibility that each superpixel block belongs to the articular disc area is obtained.

[0074] Step S004: Obtain updated grayscale values ​​of pixels in each superpixel block based on the possibility that each superpixel block belongs to the articular disc area, the grayscale value of each pixel in the superpixel block, and the grayscale value of pixels in the cluster where each pixel is located.

[0075] In order to enhance the contrast of the articular disc area, the grayscale value of the pixels in the area must be enhanced while ensuring that there is a difference in grayscale with other areas, so that the pixels in the articular disc area are enhanced without being disturbed by the surrounding areas.

[0076] The first In the superpixel block The maximum and minimum grayscale values ​​of all pixels in the cluster where the pixel is located are recorded as In the superpixel block The maximum grayscale value and the minimum grayscale value corresponding to the pixel.

[0077] This embodiment provides a calculation method for obtaining the updated grayscale value of a pixel point in a superpixel block:

[0078]

[0079] Where: Indicates In the superpixel block Updated grayscale value of pixels; Indicates In the superpixel block The minimum gray value of pixels; Indicates In the superpixel block The maximum gray value of pixels; Indicates The probability that a superpixel block belongs to the articular disc area; Indicates In the superpixel block The gray value of each pixel; represents the linear normalization function; Represents the ceiling function.

[0080] Step S005: In the temporomandibular joint grayscale image, the grayscale value of each pixel in each superpixel block is replaced with the updated grayscale value to obtain an enhanced temporomandibular joint grayscale image.

[0081] In the temporomandibular joint grayscale image, the grayscale value of the pixel in each superpixel block is replaced with the updated grayscale value to obtain an enhanced temporomandibular joint grayscale image.

[0082] By analyzing the enhanced MRI images of the temporomandibular joint, doctors can more clearly and accurately find signs of abnormalities or lesions in the soft tissue of the temporomandibular joint, thereby achieving auxiliary diagnosis and treatment of the temporomandibular joint based on medical images.

[0083] So far, the present invention is completed.

[0084] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for auxiliary diagnosis and treatment of temporomandibular joint based on medical imaging, characterized in that: The method comprises the following steps: Obtaining a grayscale image of the temporomandibular joint; A clustering operation is performed on the grayscale image of the temporomandibular joint to obtain several clusters, and a target cluster is screened out from all the clusters; in the target cluster, all continuous adjacent pixels constitute a connected domain, and the target connected domain is screened out from all the connected domains according to the area of ​​the connected domain and the gradient values ​​of all pixels on the boundary; Perform superpixel segmentation on the area except the target cluster in the grayscale image of the temporomandibular joint to obtain several superpixel blocks; divide each superpixel block into several reference sub-superpixel blocks and a neutron superpixel block; obtain the judgment parameter of each superpixel block according to the area size of the reference sub-superpixel block and the neutron superpixel block; obtain the position judgment parameter of each superpixel block according to the position of the centroid of the superpixel block and the target connected domain; obtain the possibility that each superpixel block belongs to the articular disc area according to the area of ​​the reference sub-superpixel block and the neutron superpixel block, the distance between the superpixel block and the centroid of the target connected domain, and the judgment parameter and position judgment parameter of each superpixel block; According to the possibility that each superpixel block belongs to the articular disc area, the grayscale value of each pixel in the superpixel block, and the grayscale value of the pixel in the cluster where each pixel is located, the updated grayscale value of the pixel in each superpixel block is obtained; In the temporomandibular joint grayscale image, the grayscale value of each pixel in each superpixel block is replaced with the updated grayscale value to obtain an enhanced temporomandibular joint grayscale image; The specific steps of obtaining the determination parameter of each super pixel block according to the area size of the reference sub-super pixel block and the neutron super pixel block are as follows: When the area of ​​any reference sub-superpixel block is greater than or equal to the area of ​​the neutron superpixel block, The determination parameter of the super pixel block is set to the first preset parameter T; when the area of ​​each reference sub-super pixel block in the reference sub-super pixel block is smaller than the area of ​​the neutron super pixel block, The determination parameter of each super pixel block is set to the second preset parameter E; The specific steps of obtaining the position determination parameter of each super pixel block according to the position of the centroid of the super pixel block and the target connected domain are as follows: In the grayscale image of the temporomandibular joint, a rectangular coordinate system is constructed with the vertex in the lower left corner as the origin, the horizontal rightward as the horizontal axis, and the vertical upward as the vertical axis; In the rectangular coordinate system, the value of the vertical coordinate of the pixel point is recorded as the height of the pixel point; When the height of the centroid of the superpixel block is greater than the height of the centroid of the target connected domain, The position determination parameter of each super pixel block is set to a third preset parameter R; When the height of the centroid of the superpixel block is less than or equal to the height of the centroid of the target connected domain, The position determination parameter of each super pixel block is set to a fourth preset parameter Q; The possibility that each superpixel block belongs to the articular disc region is obtained according to the area of ​​the reference sub-superpixel block and the neutron superpixel block, the distance between the superpixel block and the centroid of the target connected domain, and the judgment parameter and position judgment parameter of each superpixel block, and the specific calculation steps are as follows: Where: Indicates The probability that a superpixel block belongs to the articular disc area; Indicates The distance between the centroid of the superpixel patch and the centroid of the target connected domain; Represents the absolute value of the difference between the areas of two reference sub-superpixel blocks; Indicates The decision parameters of super pixel blocks, Indicates Position determination parameters of super pixel blocks; The method of obtaining the updated grayscale value of the pixels in each superpixel block according to the possibility that each superpixel block belongs to the articular disc area, the grayscale value of each pixel in the superpixel block, and the grayscale value of the pixels in the cluster where each pixel is located, includes the following specific steps: The first In the superpixel block The maximum and minimum grayscale values ​​of all pixels in the cluster where the pixel is located are recorded as In the superpixel block The maximum grayscale value and the minimum grayscale value corresponding to the pixel points; According to the possibility that each superpixel block belongs to the articular disc area, the grayscale value of each pixel in the superpixel block, and the maximum grayscale value and the minimum grayscale value corresponding to each pixel, the updated grayscale value of the pixel in each superpixel block is obtained; The method of obtaining the updated grayscale value of the pixel points in each superpixel block according to the possibility that each superpixel block belongs to the articular disc area, the grayscale value of each pixel point in the superpixel block, and the maximum grayscale value and the minimum grayscale value corresponding to each pixel point, includes the following specific steps: Where: Indicates In the superpixel block Updated grayscale value of pixels; Indicates In the superpixel block The minimum gray value of pixels; Indicates In the superpixel block The maximum gray value of pixels; Indicates The probability that a superpixel block belongs to the articular disc area; Indicates In the superpixel block The gray value of each pixel; represents the linear normalization function; Represents the ceiling function.

2. The method for auxiliary diagnosis and treatment of the temporomandibular joint based on medical imaging according to claim 1, characterized in that: The specific steps of selecting a target cluster from all clusters are as follows: Calculate the mean grayscale value of all pixels in each cluster, and record the cluster with the largest mean grayscale value of all pixels as the target cluster.

3. The method for auxiliary diagnosis and treatment of the temporomandibular joint based on medical imaging according to claim 1, characterized in that: The method of selecting a target connected domain from all connected domains according to the area of ​​the connected domain and the gradient values ​​of all pixels on the boundary includes the following specific steps: The connected domain of pixels that do not exist on the boundary of the temporomandibular joint grayscale image is recorded as the main connected domain; Perform convex hull detection on all main connected domains to obtain the convex hull area, calculate the area ratio of each main connected domain in the convex hull area, and record the main connected domain whose area ratio in the convex hull area is greater than a preset threshold as the reference connected domain; The target connected domain is screened out according to the gradient values ​​of all pixels on the boundary of the reference connected domain.

4. The method for auxiliary diagnosis and treatment of the temporomandibular joint based on medical imaging according to claim 3, characterized in that: The specific steps of selecting the target connected domain according to the gradient values ​​of all pixels on the boundary of the reference connected domain are as follows: The Sobel operator is used to operate the grayscale image of the temporomandibular joint to obtain the gradient value of each pixel; Calculate the mean of the gradient values ​​of all pixels on the boundary of the reference connected domain; The reference connected domain with the largest mean of the gradient values ​​of all pixels on the boundary of the reference connected domain is recorded as the target connected domain.

5. The method for auxiliary diagnosis and treatment of the temporomandibular joint based on medical imaging according to claim 1, characterized in that: The specific steps of dividing each super pixel block into a plurality of reference sub-super pixel blocks and a neutron super pixel block are as follows: The first The line connecting the two pixels with the longest distance in a superpixel block is recorded as the target straight line segment; Divide the target straight line segment into three equal segments, and the corresponding two pixel points are recorded as segmentation pixel points; The straight line that is perpendicular to the target straight line segment through the segmented pixel point is recorded as the segmentation line; Use the dividing line to separate the The super pixel block is divided into three sub-super pixel blocks, and the sub-super pixel block between two segmentation lines is recorded as the neutron super pixel block; Among the three sub-super-pixel blocks, the sub-super-pixel block that is not a neutron super-pixel block is recorded as a reference sub-super-pixel block.

Images