Needle point position identification method for thyroid fine needle puncture ultrasound image

Abstract

The invention relates to the technical field of image analysis, in particular to a needle point position identification method for thyroid fine needle puncture ultrasound images, which comprises the following steps: acquiring each frame of target ultrasound image in the current puncture process, and determining a local difference feature value corresponding to each pixel point; determining a characteristic state change index corresponding to each pixel point; screening out a target change area from each frame of target ultrasonic image, and carrying out edge detection on the target change area; screening out a target fine needle area from each frame of target ultrasonic image; determining a fine needle motion state corresponding to each frame of target ultrasonic image according to a motion change condition between the target fine needle areas in the adjacent frames of target ultrasonic images; and based on the fine needle motion state corresponding to each frame of target ultrasonic image, identifying the needle point position in each frame of target ultrasonic image. According to the method, the target ultrasonic image is analyzed, so that the needle point position is recognized, and the accuracy of needle point position recognition is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of image analysis, and in particular to a method for identifying the needle tip position of a thyroid fine needle aspiration ultrasound image. Background Art

[0002] With the development of science and technology, image recognition technology is being used more and more widely. For example, it can be applied to the needle tip position recognition in thyroid fine needle aspiration ultrasound images. Currently, the method commonly used to identify objects is to identify the object position from the image based on the grayscale value.

[0003] However, when identifying the needle tip position in thyroid fine needle aspiration ultrasound images based on grayscale values, the following technical problems often arise:

[0004] In ultrasound images, the grayscale of the needle tip is often similar to that of high-echo structures such as calcifications and fibrous septa. Therefore, when identifying the needle tip position in thyroid fine needle aspiration ultrasound images, only considering the difference in grayscale values may lead to misjudgment of the needle tip pixels, resulting in poor accuracy in needle tip position identification. Summary of the Invention

[0005] In order to solve the technical problem of poor accuracy in needle tip position recognition, the present invention proposes a needle tip position recognition method for thyroid fine needle aspiration ultrasound images.

[0006] In a first aspect, the present invention provides a method for identifying the needle tip position of a thyroid fine needle aspiration ultrasound image, the method comprising:

[0007] Obtain each frame of the target ultrasound image during the current puncture process, and determine the local difference feature value corresponding to each pixel point based on the grayscale change within a preset neighborhood corresponding to each pixel point in each frame of the target ultrasound image;

[0008] Determining a characteristic state change index corresponding to each pixel point based on a difference between a local difference characteristic value corresponding to each pixel point and a local difference characteristic value corresponding to a corresponding pixel point in a previous frame of the target ultrasound image;

[0009] According to the characteristic state change index corresponding to the pixel points in each frame of the target ultrasound image, the target change area is screened from each frame of the target ultrasound image, and the edge detection is performed on the target change area to obtain the target edge contour;

[0010] According to the shape characteristics of all target edge contours in each frame of the target ultrasound image, the target fine needle area is screened out from each frame of the target ultrasound image;

[0011] Determining the motion state of the fine needle corresponding to each frame of the target ultrasound image according to the motion change between the target fine needle areas in adjacent frames of the target ultrasound image;

[0012] Based on the motion state of the fine needle corresponding to each frame of the target ultrasound image, the needle tip position in each frame of the target ultrasound image is identified.

[0013] In combination with the first aspect above, in one possible implementation, determining the local difference feature value corresponding to each pixel point based on the grayscale change within a preset neighborhood corresponding to each pixel point in each frame of the target ultrasound image includes:

[0014] Determine a local grayscale deviation factor corresponding to each pixel point in each frame of the target ultrasound image based on the grayscale difference between each pixel point in each frame of the target ultrasound image and each pixel point in its corresponding preset neighborhood;

[0015] The mean of the local grayscale deviation factors corresponding to all pixels in a preset neighborhood corresponding to each pixel in each frame of the target ultrasound image is normalized to obtain a local difference feature value corresponding to each pixel in each frame of the target ultrasound image.

[0016] In combination with the first aspect above, in one possible implementation, determining the local grayscale deviation factor corresponding to each pixel in each frame of the target ultrasound image based on the grayscale difference between each pixel in each frame of the target ultrasound image and each pixel in its corresponding preset neighborhood includes:

[0017] Determine any frame of target ultrasound image as a marked ultrasound image, and determine any pixel point in the marked ultrasound image as a marked pixel point;

[0018] Determine the absolute value of the difference between the grayscale value corresponding to the marked pixel and the grayscale value corresponding to each pixel in its corresponding preset neighborhood as the grayscale difference, and obtain a grayscale difference set corresponding to the marked pixel;

[0019] Determine the maximum grayscale difference in the grayscale difference set corresponding to the marked pixel point as the grayscale difference peak value corresponding to the marked pixel point;

[0020] Determine the mean of all grayscale differences in the grayscale difference set corresponding to the marked pixel as the mean of the grayscale differences corresponding to the marked pixel;

[0021] The product of the grayscale difference peak value and the grayscale difference mean value corresponding to the marked pixel point is determined as the local grayscale deviation factor corresponding to the marked pixel point.

[0022] In combination with the first aspect above, in one possible implementation, determining the feature state change index corresponding to each pixel point based on the difference between the local difference feature value corresponding to each pixel point and the local difference feature value corresponding to the corresponding pixel point in the previous frame of the target ultrasound image includes:

[0023] The absolute value of the difference between the local difference eigenvalue corresponding to each pixel point in each frame of the target ultrasound image and the local difference eigenvalue corresponding to the corresponding pixel point in the previous frame of the target ultrasound image is determined as the initial state change factor corresponding to each pixel point in each frame of the target ultrasound image;

[0024] According to the initial state change factor and local grayscale deviation factor corresponding to each pixel point in each frame of the target ultrasound image, the characteristic state change index corresponding to each pixel point in each frame of the target ultrasound image is determined, wherein the initial state change factor and the local grayscale deviation factor are both positively correlated with the characteristic state change index.

[0025] In combination with the first aspect above, in one possible implementation, screening out the target change region from each frame of the target ultrasound image based on the characteristic state change index corresponding to the pixel points in each frame of the target ultrasound image includes:

[0026] Filtering out pixels whose corresponding characteristic state change index is greater than a preset change threshold from each frame of the target ultrasound image as changed pixels;

[0027] The minimum circumscribed rectangle of the area formed by all the changed pixels in each frame of the target ultrasound image is determined as the target changed area.

[0028] In combination with the first aspect above, in one possible implementation, screening out the target fine needle region from each frame of the target ultrasound image based on the shape characteristics of all target edge contours in each frame of the target ultrasound image includes:

[0029] Determine the local contour direction vector corresponding to each edge pixel point on each target edge contour according to each edge pixel point and its adjacent edge pixel points;

[0030] The absolute value of the cosine similarity between the local contour direction vectors corresponding to each edge pixel point on each target edge contour and any one of its adjacent edge pixel points is determined as the direction change similarity corresponding to each edge pixel point on each target edge contour;

[0031] According to the difference between the similarities of the direction changes corresponding to all adjacent edge pixels on each target edge contour, the possible indicators of the fine needle shape corresponding to each target edge contour are determined;

[0032] Filter out the target edge contour with the largest possible index of the corresponding fine needle shape from each frame of the target ultrasound image as the candidate fine needle contour;

[0033] The edge pixels on the candidate outline of the fine needle in each frame of the target ultrasound image are used as initial seed points for region growing to obtain the target subregion;

[0034] The union of the inner area of the thin needle candidate contour and the target sub-area to which the thin needle candidate contour belongs is determined as the target thin needle area.

[0035] In combination with the first aspect above, in one possible implementation, determining the local contour direction vector corresponding to each edge pixel point on each target edge contour based on each edge pixel point and its adjacent edge pixel points includes:

[0036] Determine any target edge contour as a marked edge contour, and determine any edge pixel point on the marked edge contour as a marked edge point;

[0037] Determine any one edge pixel point among two edge pixel points adjacent to the marked edge point on the marked edge contour as a reference edge point;

[0038] Based on the marked edge point and the reference edge point, a local contour direction vector corresponding to the marked edge point is constructed, wherein the modulus of the local contour direction vector corresponding to the marked edge point is equal to the distance between the marked edge point and the reference edge point; and the direction of the local contour direction vector corresponding to the marked edge point is the direction from the marked edge point to the reference edge point.

[0039] In conjunction with the first aspect above, in one possible implementation, determining the possible indicator of the fine needle shape corresponding to each target edge contour based on the difference between the similarities of the direction changes corresponding to all adjacent edge pixels on each target edge contour includes:

[0040] The absolute value of the difference between the direction change similarities corresponding to each adjacent edge pixel point on each target edge contour is determined as the change similarity difference, and the change similarity difference set corresponding to each target edge contour is obtained;

[0041] The mean of all the change similarity differences in the change similarity difference set corresponding to each target edge contour is determined as the overall similarity difference corresponding to each target edge contour;

[0042] According to the overall similarity difference corresponding to each target edge contour, the possible index of the fine needle shape corresponding to each target edge contour is determined, wherein the overall similarity difference is negatively correlated with the possible index of the fine needle shape.

[0043] In combination with the first aspect above, in one possible implementation, determining the fine needle motion state corresponding to each frame of the target ultrasound image based on the motion change between the target fine needle regions in adjacent frames of the target ultrasound image includes:

[0044] Filter out the pixel point with the largest corresponding characteristic state change index from the target fine needle area in each frame of the target ultrasound image as the needle tip candidate point;

[0045] Screening out the pixel point with the largest grayscale difference from the candidate needle tip point in a preset neighborhood corresponding to the candidate needle tip point in each frame of the target ultrasound image as the reference point corresponding to the candidate needle tip point in each frame of the target ultrasound image;

[0046] Constructing a reference direction vector corresponding to each frame of the target ultrasound image based on the candidate needle tip point and its corresponding reference point in each frame of the target ultrasound image, wherein the modulus of the reference direction vector is equal to the distance between the candidate needle tip point and its corresponding reference point, and the direction of the reference direction vector is the direction from the candidate needle tip point to its corresponding reference point;

[0047] If the cosine similarity between the reference direction vectors corresponding to adjacent target ultrasound images is greater than a preset similarity threshold, it is determined that the fine needle motion states corresponding to the adjacent target ultrasound images are the same;

[0048] The target ultrasound images of consecutive frames with the same needle motion state are formed into a target ultrasound image group to obtain a target ultrasound image group sequence;

[0049] setting the fine needle movement states corresponding to all target ultrasound images in the target ultrasound image groups with odd numbers in the target ultrasound image group sequence as fine needle insertion states;

[0050] The fine needle motion states corresponding to all target ultrasound images in the target ultrasound image groups with even numbers in the target ultrasound image group sequence are set as the fine needle withdrawal state.

[0051] In combination with the first aspect above, in one possible implementation, identifying the needle tip position in each frame of the target ultrasound image based on the fine needle motion state corresponding to each frame of the target ultrasound image includes:

[0052] If the fine needle motion state corresponding to the target ultrasound image is the fine needle insertion state, the position of the needle tip candidate point in the target ultrasound image is determined as the needle tip position in the target ultrasound image;

[0053] If the fine needle motion state corresponding to the target ultrasound image is the fine needle withdrawal state, the position of the target fine needle region to which the needle tip candidate point belongs in the target ultrasound image is determined as the needle tip position in the target ultrasound image.

[0054] In a second aspect, the present invention provides a needle tip position recognition system for thyroid fine needle aspiration ultrasound images, the system comprising:

[0055] An acquisition and determination module is used to acquire each frame of the target ultrasound image during the current puncture process and determine the local difference feature value corresponding to each pixel point based on the grayscale change within a preset neighborhood corresponding to each pixel point in each frame of the target ultrasound image;

[0056] a feature state change index determination module, configured to determine a feature state change index corresponding to each pixel point based on a difference between a local difference feature value corresponding to each pixel point and a local difference feature value corresponding to a corresponding pixel point in a previous frame of target ultrasound image;

[0057] A screening and edge detection module is used to screen out target change areas from each frame of the target ultrasound image based on the characteristic state change index corresponding to the pixel points in each frame of the target ultrasound image, and perform edge detection on the target change area to obtain the target edge contour;

[0058] A region screening module is used to screen out the target fine needle region from each frame of the target ultrasound image based on the shape characteristics of all target edge contours in each frame of the target ultrasound image;

[0059] A fine needle motion state determination module is used to determine the fine needle motion state corresponding to each frame of the target ultrasound image based on the motion changes between the target fine needle areas in adjacent frames of the target ultrasound image;

[0060] The needle tip position recognition module is used to recognize the needle tip position in each frame of the target ultrasound image based on the fine needle motion state corresponding to each frame of the target ultrasound image.

[0061] In a third aspect, a server is provided, comprising a memory and a processor. The memory is configured to store executable program code, and the processor is configured to call and execute the executable program code from the memory, so that the device executes the method of the first aspect or any possible implementation of the first aspect.

[0062] In a fourth aspect, a computer program product is provided, comprising: a computer program code, which, when executed on a computer, enables the computer to execute the method in the first aspect or any possible implementation of the first aspect.

[0063] In a fifth aspect, a computer-readable storage medium is provided, which stores a computer program code. When the computer program code runs on a computer, the computer executes the method in the above-mentioned first aspect or any possible implementation of the first aspect.

[0064] The present invention has the following beneficial effects:

[0065] The present invention provides a needle tip position identification method for thyroid fine needle aspiration ultrasound images. By analyzing the target ultrasound image, the needle tip position is identified, thereby resolving the technical problem of poor accuracy in needle tip position identification and improving the accuracy of needle tip position identification. When performing needle tip position identification, the present invention comprehensively considers multiple factors related to needle tip position identification, such as local difference eigenvalues and characteristic state change indicators, thereby more accurately identifying the target fine needle region representing the fine needle to which the needle tip belongs. Furthermore, the present invention comprehensively considers the motion changes between the target fine needle regions in adjacent frames of target ultrasound images, quantifies the fine needle motion state corresponding to each frame of the target ultrasound image, thereby achieving needle tip position identification and, to a certain extent, improving the accuracy of needle tip position identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. 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 any creative work.

[0067] Figure 1 This is a flow chart of a method for identifying the needle tip position of a thyroid fine needle aspiration ultrasound image according to the present invention;

[0068] Figure 2 This is a schematic diagram of the structure of a needle tip position recognition system for thyroid fine needle aspiration ultrasound images according to the present invention;

[0069] Figure 3 The figure is a structural diagram of a computer device of the present invention. DETAILED DESCRIPTION

[0070] To further illustrate the technical means and effects employed by the present invention to achieve its intended objectives, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementations, structures, features, and effects of the technical solutions proposed by the present invention. In the following description, references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0071] 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.

[0072] refer to Figure 1, shows the process of some embodiments of a method for identifying the needle tip position of a thyroid fine needle aspiration ultrasound image according to the present invention. The method for identifying the needle tip position of a thyroid fine needle aspiration ultrasound image comprises the following steps:

[0073] Step S1: Obtain each frame of the target ultrasound image during the current puncture process, and determine the local difference feature value corresponding to each pixel point based on the grayscale change within a preset neighborhood corresponding to each pixel point in each frame of the target ultrasound image.

[0074] Among them, the current puncture process can be a thyroid fine needle aspiration process that is currently being carried out. Fine needle aspiration (FNA) of the thyroid is a diagnostic examination method commonly used to evaluate thyroid nodules or masses. This method uses a fine needle to extract a small amount of cell samples from the thyroid nodule or mass for examination under a microscope to help the doctor determine the nature of the nodule. A complete thyroid fine needle aspiration process can be a process in which the doctor uses ultrasound guidance to pass a thin needle through the skin, enter the thyroid nodule, and completely pull out the needle. The target ultrasound image can be an ultrasound image of the thyroid nodule collected during the thyroid fine needle aspiration process. The preset neighborhood can be a pre-set neighborhood. For example, the preset neighborhood can be an eight-neighborhood neighborhood.

[0075] It should be noted that when guiding fine needle aspiration of a patient's thyroid nodule, one or two medical staff are often required to hold a linear array ultrasound probe to obtain an ultrasound image of the thyroid nodule area while holding the puncture needle to operate. During the puncture process of the patient, multiple guidance is often required to ensure that the thyroid nodule can be accurately punctured and the sample can be obtained. Therefore, when performing thyroid nodule puncture on a patient, it is often necessary to constantly locate the needle tip position to more accurately puncture the nodule and obtain the sample.

[0076] As an example, this step may include the following steps:

[0077] The first step is to obtain each frame of the target ultrasound image during the current puncture process.

[0078] For example, a linear array ultrasound probe may be used to collect ultrasound images of the thyroid nodule during the current puncture process, and each frame of the ultrasound image collected during the current puncture process may be recorded as a target ultrasound image.

[0079] Among them, the first frame target ultrasound image can be an ultrasound image collected at the beginning of the current puncture process. At this time, when the first frame target ultrasound image is collected, the fine needle is often not used to pass through the skin. Therefore, there is often no fine needle area in the first frame target ultrasound image, which means that there is often no need to identify the needle tip position of the first frame target ultrasound image subsequently.

[0080] It should be noted that by using a linear array probe to detect the patient's thyroid nodule area, an ultrasound image of the patient's thyroid nodule area can be obtained.

[0081] The second step, based on the grayscale difference between each pixel in each frame of the target ultrasound image and each pixel in its corresponding preset neighborhood, determines the local grayscale deviation factor corresponding to each pixel in each frame of the target ultrasound image, which may include the following sub-steps:

[0082] In the first sub-step, any frame of the target ultrasound image is determined as a marked ultrasound image, and any pixel point in the marked ultrasound image is determined as a marked pixel point.

[0083] In the second sub-step, the absolute value of the difference between the grayscale value corresponding to the marked pixel point and the grayscale value corresponding to each pixel point in its corresponding preset neighborhood is determined as the grayscale difference, and a grayscale difference set corresponding to the marked pixel point is obtained.

[0084] The grayscale difference set corresponding to the marked pixel may include: the absolute value of the difference between the grayscale value corresponding to the marked pixel and the grayscale values corresponding to each pixel in the corresponding preset neighborhood.

[0085] In a third sub-step, the maximum grayscale difference in the grayscale difference set corresponding to the marked pixel point is determined as the grayscale difference peak value corresponding to the marked pixel point.

[0086] In a fourth sub-step, the mean of all grayscale differences in the grayscale difference set corresponding to the marked pixel point is determined as the mean grayscale difference corresponding to the marked pixel point.

[0087] In a fifth sub-step, the product of the grayscale difference peak value and the grayscale difference mean value corresponding to the marked pixel point is determined as the local grayscale deviation factor corresponding to the marked pixel point.

[0088] For example, the formula for determining the local grayscale deviation factor corresponding to a pixel point in the target ultrasound image can be:

[0089] δ i,j =MP i,j ×μP i,j ; Among them, δ i,j is the local grayscale deviation factor corresponding to the jth pixel in the target ultrasound image of the i-th frame. i is the serial number of the target ultrasound image. j is the serial number of the pixel in the target ultrasound image of the i-th frame. i,j It is the grayscale difference peak value corresponding to the jth pixel in the target ultrasound image of the i-th frame, that is, the maximum grayscale difference in the grayscale difference set corresponding to the jth pixel in the target ultrasound image of the i-th frame. i,jIt is the mean grayscale difference corresponding to the j-th pixel in the target ultrasound image of the i-th frame, that is, the mean of all grayscale differences in the grayscale difference set corresponding to the j-th pixel in the target ultrasound image of the i-th frame.

[0090] It should be noted that during the puncture process, the grayscale of the area where the needle is inserted into the patient is often different from the grayscale of the human tissue. Therefore, the grayscale corresponding to the pixel point of the needle is often different from the grayscale in its neighborhood. i,j When MP is larger, it often indicates that the maximum grayscale difference between the jth pixel and its neighborhood is larger. i,j Can be used as μP i,j When μP i,j The larger the value, the greater the overall grayscale difference between the jth pixel and its neighborhood. i,j It can characterize the grayscale distribution around the j-th pixel. The larger its value is, the greater the grayscale difference between the j-th pixel and its neighborhood is, which often means that the j-th pixel is more likely to be a fine needle pixel.

[0091] In the third step, the mean of the local grayscale deviation factors corresponding to all pixels in the preset neighborhood corresponding to each pixel in each frame of the target ultrasound image is normalized to obtain the local difference feature value corresponding to each pixel in each frame of the target ultrasound image.

[0092] For example, the formula for determining the local difference eigenvalue corresponding to the pixel point in the target ultrasound image may be:

[0093] in, is the local difference eigenvalue corresponding to the jth pixel in the target ultrasound image of the i-th frame. i is the serial number of the target ultrasound image. j is the serial number of the pixel in the target ultrasound image of the i-th frame. norm() is the normalization function. n ij is the number of pixels in the preset neighborhood corresponding to the jth pixel in the target ultrasound image of the i-th frame. a is the serial number of the pixel in the preset neighborhood corresponding to the jth pixel in the target ultrasound image of the i-th frame. i,j is the local grayscale deviation factor corresponding to the jth pixel in the i-th frame target ultrasound image.

[0094] It should be noted that It can represent the grayscale change around the j-th pixel. The larger its value is, the greater the grayscale change around the j-th pixel is.

[0095] Step S2: determining a characteristic state change index corresponding to each pixel point based on the difference between the local difference characteristic value corresponding to each pixel point and the local difference characteristic value corresponding to the corresponding pixel point in the previous frame of the target ultrasound image.

[0096] Among them, any pixel point is recorded as a temporary pixel point, and the corresponding pixel point in the previous frame target ultrasound image of the temporary pixel point can be: the pixel point in the previous frame target ultrasound image to which the temporary pixel point belongs, which has the same position as the temporary pixel point.

[0097] As an example, this step may include the following steps:

[0098] In the first step, the absolute value of the difference between the local difference eigenvalue corresponding to each pixel point in each frame of the target ultrasound image and the local difference eigenvalue corresponding to the corresponding pixel point in the previous frame of the target ultrasound image is determined as the initial state change factor corresponding to each pixel point in each frame of the target ultrasound image.

[0099] In the second step, the characteristic state change index corresponding to each pixel point in each frame of the target ultrasound image is determined according to the initial state change factor and the local grayscale deviation factor corresponding to each pixel point in each frame of the target ultrasound image.

[0100] Among them, the initial state change factor and the local grayscale deviation factor can both be positively correlated with the characteristic state change index.

[0101] For example, the formula for determining the feature state change index corresponding to a pixel point can be:

[0102] Among them, θ i,j is the feature state change index corresponding to the jth pixel in the target ultrasound image of the i-th frame. i is the serial number of the target ultrasound image. j is the serial number of the pixel in the target ultrasound image of the i-th frame. δ i,j is the local grayscale deviation factor corresponding to the jth pixel in the target ultrasound image of the i-th frame. || is the absolute value function. is the local difference eigenvalue corresponding to the jth pixel in the target ultrasound image of the i-th frame. is the local difference eigenvalue corresponding to the jth pixel in the target ultrasound image of frame i-1. The jth pixel in the target ultrasound image of frame i-1 corresponds to the pixel in the target ultrasound image of frame i preceding the jth pixel in the target ultrasound image of frame i. γ is a pre-set factor greater than 0, primarily used to prevent the denominator from being zero. For example, γ can be 0.001. is the initial state variation factor corresponding to the jth pixel point in the target ultrasound image of the i-th frame.

[0103] It should be noted that during the puncture process, the grayscale of the area where the fine needle is inserted into the patient often differs from the grayscale of the human tissue. Therefore, the grayscale corresponding to the fine needle pixel often differs from the grayscale in its neighborhood. In addition, the fine needle traction often causes deviations between the grayscales of pixels at the same position in adjacent frames of ultrasound images. Therefore, the fine needle area can be identified by comparing the grayscale changes between pixels at the same position in adjacent frames of ultrasound images. i,j It can characterize the grayscale distribution around the j-th pixel. The larger its value is, the greater the grayscale difference between the j-th pixel and its neighborhood is, which often means that the j-th pixel is more likely to be a fine needle pixel. When θ is larger, it often means that the surrounding grayscale changes corresponding to the j-th pixel at the same position in the target ultrasound image of adjacent frames are more dissimilar, and it often means that the j-th pixel is more likely to be a fine needle pixel. i,j The larger the value is, the more likely the jth pixel in the target ultrasound image of the i-th frame is to be a fine needle pixel.

[0104] Step S3 , based on the characteristic state change index corresponding to the pixel points in each frame of the target ultrasound image, a target change area is screened out from each frame of the target ultrasound image, and edge detection is performed on the target change area to obtain a target edge contour.

[0105] As an example, this step may include the following steps:

[0106] In the first step, pixels whose corresponding characteristic state change index is greater than a preset change threshold are screened out from each frame of the target ultrasound image as changed pixels.

[0107] The preset change threshold may be a preset minimum characteristic state change index required to determine that a pixel point has changed. For example, the preset change threshold may be 0.3.

[0108] In the second step, the minimum bounding rectangle of the area formed by all the changed pixels in each frame of the target ultrasound image is determined as the target changed area.

[0109] The third step is to perform edge detection on the target change area to obtain the target edge contour.

[0110] For example, the Sobel bidirectional edge detection algorithm may be used to perform edge detection on the target change region, and the edge contour obtained at this time is recorded as the target edge contour.

[0111] Step S4: Filter out the target fine needle region from each frame of the target ultrasound image according to the shape characteristics of all target edge contours in each frame of the target ultrasound image.

[0112] It should be noted that, since the shape of the fine needle region often has a certain uniqueness, analyzing the shape characteristics of the target edge contour can facilitate the subsequent screening of the target fine needle region that represents the fine needle.

[0113] As an example, this step may include the following steps:

[0114] The first step, based on each edge pixel point on each target edge contour and its adjacent edge pixel points, determines the local contour direction vector corresponding to each edge pixel point on each target edge contour, which may include the following sub-steps:

[0115] In the first sub-step, any target edge contour is determined as a marked edge contour, and any edge pixel point on the marked edge contour is determined as a marked edge point.

[0116] In the second sub-step, any one of the two edge pixel points adjacent to the marked edge point on the marked edge contour is determined as a reference edge point.

[0117] It should be noted that there are often two edge pixel points adjacent to the marked edge point on the marked edge contour, and the reference edge point is one of them.

[0118] The third sub-step is to construct a local contour direction vector corresponding to the marked edge point according to the marked edge point and the reference edge point.

[0119] The modulus of the local contour direction vector corresponding to the marked edge point can be equal to the distance between the marked edge point and the reference edge point. The direction of the local contour direction vector corresponding to the marked edge point can be the direction from the marked edge point to the reference edge point. The local contour direction vector corresponding to the marked edge point can, to a certain extent, represent the vector of the line connecting the marked edge point and the adjacent edge pixel point, and can also be referred to as the line vector corresponding to the marked edge point.

[0120] In the second step, the absolute value of the cosine similarity between the local contour direction vector corresponding to each edge pixel point on each target edge contour and any of its adjacent edge pixel points is determined as the direction change similarity corresponding to each edge pixel point on each target edge contour.

[0121] For example, the absolute value of the cosine similarity between the local contour direction vector corresponding to the marked edge point and the local contour direction vector corresponding to the reference edge point may be determined as the direction change similarity corresponding to the marked edge point.

[0122] It should be noted that since fine needles often have a linear shape, the angle between the line vectors connecting adjacent edge pixels on a fine needle is often 0° or 180°. Therefore, the cosine similarity between the line vectors corresponding to adjacent fine needle edge pixels is often 1 or -1, and the absolute value of the cosine similarity between the line vectors corresponding to adjacent fine needle edge pixels is often 1. In other words, the similarity of the directional changes corresponding to the fine needle edge pixels is often the same, often 1.

[0123] The third step is to determine the possible indicators of the fine needle shape corresponding to each target edge contour based on the difference between the similarities of the direction changes corresponding to all adjacent edge pixels on each target edge contour, which may include the following sub-steps:

[0124] In the first sub-step, the absolute value of the difference between the direction change similarities corresponding to each adjacent edge pixel point on each target edge contour is determined as the change similarity difference, and a change similarity difference set corresponding to each target edge contour is obtained.

[0125] The change similarity difference set corresponding to the target edge contour may include: the absolute values of the differences between the direction change similarities corresponding to all adjacent edge pixels on the target edge contour.

[0126] In the second sub-step, the mean of all the change similarity differences in the change similarity difference set corresponding to each target edge contour is determined as the overall similarity difference corresponding to each target edge contour.

[0127] In the third sub-step, the possible indicators of the fine needle shape corresponding to each target edge contour are determined based on the overall similarity difference corresponding to each target edge contour.

[0128] Among them, the overall similarity difference can be negatively correlated with the possible indicators of fine needle shape.

[0129] For example, the formula for determining the possible indicators of the fine needle shape corresponding to the target edge contour can be:

[0130] Among them, w t is the possible index of the fine needle shape corresponding to the t-th target edge contour. t is the serial number of the target edge contour. exp() is the natural exponential function. m t is the number of edge pixels on the t-th target edge contour. b is the serial number of the edge pixel on the t-th target edge contour. || is the absolute value function. CS t,b CS is the similarity of the direction change corresponding to the bth edge pixel on the tth target edge contour. t,b+1 It is the similarity of the direction change corresponding to the b+1th edge pixel point on the tth target edge contour. is the overall similarity difference corresponding to the edge contour of the t-th target. |CS t,b -CS t,b+1 |It is the change similarity difference.

[0131] It should be noted that the similarity of direction changes corresponding to different needle edge pixels is often the same. t,b -CS t,b+1 The smaller | is, the more similar the directional changes of different edge pixels on the edge contour of the t-th target are. This often means that the edge contour of the t-th target is more likely to be linear, and the edge contour of the t-th target is more likely to be a fine needle contour. Therefore, when w t The larger it is, the more likely the t-th target edge contour is to be a fine needle contour.

[0132] In the fourth step, the target edge contour with the largest possible index of the corresponding fine needle shape is screened out from each frame of the target ultrasound image as the candidate fine needle contour.

[0133] In the fifth step, the edge pixel points on the candidate outline of the fine needle in each frame of the target ultrasound image are used as initial seed points to perform region growing to obtain the target sub-region.

[0134] The target sub-region may be a region obtained after region growing.

[0135] For example, the edge pixel points on the candidate outline of the fine needle can be used as initial seed points, and based on the grayscale values corresponding to the pixel points in the target ultrasound image, region growth is performed on the pixel points in the target ultrasound image, and each region obtained by region growth is recorded as a target sub-region.

[0136] It should be noted that further modifying the candidate outline of the fine needle through region growing can often facilitate the subsequent acquisition of the target fine needle region representing the real fine needle.

[0137] In the sixth step, the union of the inner area of the thin needle candidate contour and the target sub-area to which the thin needle candidate contour belongs is determined as the target thin needle area.

[0138] Among them, the inner area of the thin needle candidate contour is the area surrounded by the thin needle candidate contour.

[0139] Step S5, determining the fine needle motion state corresponding to each frame of the target ultrasound image according to the motion change between the target fine needle regions in adjacent frames of the target ultrasound image.

[0140] As an example, this step may include the following steps:

[0141] In the first step, the pixel point with the largest corresponding characteristic state change index in the target fine needle area in each frame of the target ultrasound image is selected as the candidate needle tip point.

[0142] In the second step, the pixel point with the largest grayscale difference from the needle tip candidate point in each frame of the target ultrasound image is selected from the preset neighborhood corresponding to the needle tip candidate point in each frame of the target ultrasound image as the reference point corresponding to the needle tip candidate point in each frame of the target ultrasound image.

[0143] The reference point may be a pixel point with the largest grayscale difference between the pinpoint candidate point and the pinpoint candidate point within a preset neighborhood corresponding to the pinpoint candidate point.

[0144] The third step is to construct a reference direction vector corresponding to each frame of the target ultrasound image based on the needle tip candidate point in each frame of the target ultrasound image and its corresponding reference point.

[0145] The magnitude of the reference direction vector may be equal to the distance between the needle tip candidate point and its corresponding reference point. The direction of the reference direction vector may be the direction from the needle tip candidate point to its corresponding reference point.

[0146] In the fourth step, if the cosine similarity between the reference direction vectors corresponding to adjacent target ultrasound images is greater than a preset similarity threshold, it is determined that the fine needle motion states corresponding to the adjacent target ultrasound images are the same.

[0147] The preset similarity threshold may be 0.

[0148] It should be noted that if the cosine similarity between the reference direction vectors corresponding to two adjacent frames of target ultrasound images is greater than 0, it often means that the angle between the reference direction vectors corresponding to the two frames of target ultrasound images is less than 90°, which often means that the moving direction of the needle tip candidate point in the two frames of target ultrasound images may not change significantly, and often means that the motion state of the fine needle in the two frames of target ultrasound images may not change.

[0149] In the fifth step, if the cosine similarity between the reference direction vectors corresponding to adjacent target ultrasound images is less than or equal to a preset similarity threshold, it is determined that the fine needle motion states corresponding to the adjacent target ultrasound images are different.

[0150] In the sixth step, the target ultrasound images of consecutive frames with the same needle motion state are combined into a target ultrasound image group to obtain a target ultrasound image group sequence.

[0151] The target ultrasound image group often does not include the first frame of the target ultrasound image.

[0152] For example, if a total of 8 frames of target ultrasound images are collected during the current puncture process, and these 8 frames of target ultrasound images are the first image, the second image, the third image, the fourth image, the fifth image, the sixth image, the seventh image and the eighth image in sequence; after calculation, the second image and the third image have the same corresponding fine needle motion state; the third image and the fourth image have the same corresponding fine needle motion state; the fourth image and the fifth image have different corresponding fine needle motion states; the fifth image and the sixth image have the same corresponding fine needle motion state; the sixth image and the seventh image have different corresponding fine needle motion states; the seventh image and the eighth image have different corresponding fine needle motion states. If the corresponding fine needle motion states are the same, three target ultrasound image groups can be obtained at this time. These three target ultrasound image groups are {second image, third image, fourth image}, {fifth image, sixth image} and {seventh image, eighth image}, and the sequence number of {second image, third image, fourth image} in the target ultrasound image group sequence is 1, which is an odd sequence number; the sequence number of {fifth image, sixth image} in the target ultrasound image group sequence is 2, which is an even sequence number; the sequence number of {seventh image, eighth image} in the target ultrasound image group sequence is 3, which is an odd sequence number.

[0153] In the seventh step, the fine needle motion states corresponding to all target ultrasound images in the target ultrasound image groups with odd numbers in the target ultrasound image group sequence are set as the fine needle insertion state.

[0154] In the eighth step, the fine needle movement states corresponding to all target ultrasound images in the target ultrasound image groups with even numbers in the target ultrasound image group sequence are set to the fine needle withdrawal state.

[0155] It should be noted that during the initial stages of a thyroid fine needle aspiration procedure, the needle tip is often inserted into the patient's skin. Therefore, the needle tip in ultrasound images captured during this initial stage is often in an inserted state. The target ultrasound images in the first target ultrasound image group often belong to the initial stages of the current puncture procedure, indicating that the needle tip in each target ultrasound image in the first target ultrasound image group is in an inserted state, meaning that the needle tip at this stage often penetrates deeper. The needle tip in the second target ultrasound image group is similar to the needle tip in the first target ultrasound image group. Since the needle tip in the first target ultrasound image group was in an inserted state, the needle tip in the second target ultrasound image group is often in a withdrawn state, meaning that the needle tip is slightly withdrawn. This suggests that the needle tip was inserted incorrectly at the previous moment, and the needle tip position was adjusted to withdraw it slightly to allow it to penetrate the correct position later. Similarly, the needle tip in odd-numbered target ultrasound image groups is often in an inserted state, while the needle tip in even-numbered target ultrasound image groups is often withdrawn.

[0156] Step S6: identifying the needle tip position in each frame of the target ultrasound image based on the fine needle motion state corresponding to each frame of the target ultrasound image.

[0157] As an example, this step may include the following steps:

[0158] In the first step, if the fine needle motion state corresponding to the target ultrasound image is the fine needle insertion state, the position of the needle tip candidate point in the target ultrasound image is determined as the needle tip position in the target ultrasound image.

[0159] It should be noted that during needle penetration, the subcutaneous tissue immediately adjacent to the needle often undergoes changes due to friction from the needle's surface. Furthermore, the needle tip often needs to disrupt existing subcutaneous tissue to facilitate entry. Therefore, the subcutaneous tissue at the needle tip undergoes more significant changes. Consequently, the characteristic state at the needle tip often changes the most compared to other locations of the needle. In other words, during needle penetration, the candidate needle tip point often represents the actual needle tip, and the needle tip position at this point is the location of the candidate needle tip point. Furthermore, the direction of the reference direction vector corresponding to the target ultrasound image often corresponds to the needle tip penetration direction in that target ultrasound image.

[0160] In the second step, if the fine needle motion state corresponding to the target ultrasound image is the fine needle withdrawal state, the position of the target fine needle region to which the needle tip candidate point belongs in the target ultrasound image is determined as the needle tip position in the target ultrasound image.

[0161] It should be noted that during needle tip removal, the damage to subcutaneous tissue caused by the needle tip is often less than during insertion. At this time, the characteristic state changes near the needle tip are often similar. Therefore, the candidate needle tip point may not be the actual needle tip, but rather a location near the needle tip. Therefore, it is possible to mark the target fine needle region representing the needle to which the needle tip belongs. This facilitates observation and positioning of the target fine needle region. Furthermore, the opposite direction of the reference direction vector corresponding to the target ultrasound image is often the needle tip removal direction in that target ultrasound image.

[0162] refer to Figure 2 Based on the same inventive concept as the above-mentioned method embodiment, the present invention provides a system for identifying the needle tip position of a thyroid fine needle aspiration ultrasound image. The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, the steps of a method for identifying the needle tip position of a thyroid fine needle aspiration ultrasound image may specifically include:

[0163] The acquisition and determination module 201 is used to acquire each frame of the target ultrasound image during the current puncture process and determine the local difference feature value corresponding to each pixel point based on the grayscale change within a preset neighborhood corresponding to each pixel point in each frame of the target ultrasound image;

[0164] A feature state change index determining module 202 is configured to determine a feature state change index corresponding to each pixel point based on a difference between a local difference feature value corresponding to each pixel point and a local difference feature value corresponding to a corresponding pixel point in a previous frame of the target ultrasound image;

[0165] The screening and edge detection module 203 is used to screen out the target change area from each frame of the target ultrasound image according to the characteristic state change index corresponding to the pixel points in each frame of the target ultrasound image, and perform edge detection on the target change area to obtain the target edge contour;

[0166] The region screening module 204 is configured to screen out the target fine needle region from each frame of the target ultrasound image according to the shape characteristics of all target edge contours in each frame of the target ultrasound image;

[0167] The fine needle motion state determination module 205 is used to determine the fine needle motion state corresponding to each frame of the target ultrasound image according to the motion change between the target fine needle areas in adjacent frames of the target ultrasound image;

[0168] The needle tip position recognition module 206 is configured to recognize the needle tip position in each frame of the target ultrasound image based on the fine needle motion state corresponding to each frame of the target ultrasound image.

[0169] Figure 3 FIG. 1 is a schematic diagram of the structure of a computer device provided by an embodiment of the present invention. For example, Figure 3 As shown, the computer device 300 includes: a memory 301, a processor 302, and a computer program 303 stored in the memory 301 and running on the processor 302, wherein when the processor 302 executes the computer program 303, the computer device can execute any one of the needle tip position recognition methods for thyroid fine needle aspiration ultrasound images introduced above.

[0170] Based on the same inventive concept as the above-described method embodiments, the present invention provides a server comprising a memory and a processor. The memory is configured to store executable program code, and the processor is configured to retrieve and execute the executable program code from the memory, thereby enabling the device to perform any of the above-described methods for identifying the needle tip position in thyroid fine needle aspiration ultrasound images.

[0171] Based on the same inventive concept as the above-mentioned method embodiment, the present invention provides a computer program product, which includes: computer program code, which, when running on a computer, enables the computer to execute any one of the above-mentioned methods for identifying the needle tip position of thyroid fine needle aspiration ultrasound images.

[0172] Based on the same inventive concept as the above-mentioned method embodiment, the present invention provides a computer-readable storage medium, which stores computer program code. When the computer program code runs on a computer, the computer executes any one of the above-mentioned methods for identifying the needle tip position of thyroid fine needle aspiration ultrasound images.

[0173] In summary, the present invention comprehensively considers multiple factors related to needle tip position recognition when performing needle tip position recognition, such as local difference eigenvalues and characteristic state change indicators, so as to more accurately identify the target fine needle area that represents the fine needle to which the needle tip belongs, and comprehensively considers the motion changes between the target fine needle areas in adjacent frame target ultrasound images, quantifies the fine needle motion state corresponding to each frame target ultrasound image, thereby realizing needle tip position recognition and improving the accuracy of needle tip position recognition to a certain extent.

[0174] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention, and should all be included in the scope of protection of the present invention.

Claims

1. A method for identifying the needle tip position of a thyroid fine needle aspiration ultrasound image, characterized in that: The following steps are involved: Obtain each frame of the target ultrasound image during the current puncture process, and determine the local difference feature value corresponding to each pixel point based on the grayscale change within a preset neighborhood corresponding to each pixel point in each frame of the target ultrasound image; Determining a characteristic state change index corresponding to each pixel point based on a difference between a local difference characteristic value corresponding to each pixel point and a local difference characteristic value corresponding to a corresponding pixel point in a previous frame of the target ultrasound image; According to the characteristic state change index corresponding to the pixel points in each frame of the target ultrasound image, the target change area is screened from each frame of the target ultrasound image, and the edge detection is performed on the target change area to obtain the target edge contour; According to the shape characteristics of all target edge contours in each frame of the target ultrasound image, the target fine needle area is screened out from each frame of the target ultrasound image; Determining the motion state of the fine needle corresponding to each frame of the target ultrasound image according to the motion change between the target fine needle areas in adjacent frames of the target ultrasound image; Based on the motion state of the fine needle corresponding to each frame of the target ultrasound image, the needle tip position in each frame of the target ultrasound image is identified.

2. The method for identifying the needle tip position of a thyroid fine needle aspiration ultrasound image according to claim 1, characterized in that: The determining of the local difference feature value corresponding to each pixel point based on the grayscale change in the preset neighborhood corresponding to each pixel point in each frame of the target ultrasound image includes: Determine a local grayscale deviation factor corresponding to each pixel point in each frame of the target ultrasound image based on the grayscale difference between each pixel point in each frame of the target ultrasound image and each pixel point in its corresponding preset neighborhood; The mean of the local grayscale deviation factors corresponding to all pixels in a preset neighborhood corresponding to each pixel in each frame of the target ultrasound image is normalized to obtain a local difference feature value corresponding to each pixel in each frame of the target ultrasound image.

3. The method for identifying the needle tip position of a thyroid fine needle aspiration ultrasound image according to claim 2, characterized in that: Determining the local grayscale deviation factor corresponding to each pixel point in each frame of the target ultrasound image according to the grayscale difference between each pixel point in each frame of the target ultrasound image and each pixel point in the corresponding preset neighborhood includes: Determine any frame of target ultrasound image as a marked ultrasound image, and determine any pixel point in the marked ultrasound image as a marked pixel point; Determine the absolute value of the difference between the grayscale value corresponding to the marked pixel and the grayscale value corresponding to each pixel in its corresponding preset neighborhood as the grayscale difference, and obtain a grayscale difference set corresponding to the marked pixel; Determine the maximum grayscale difference in the grayscale difference set corresponding to the marked pixel point as the grayscale difference peak value corresponding to the marked pixel point; Determine the mean of all grayscale differences in the grayscale difference set corresponding to the marked pixel as the mean of the grayscale differences corresponding to the marked pixel; The product of the grayscale difference peak value and the grayscale difference mean value corresponding to the marked pixel point is determined as the local grayscale deviation factor corresponding to the marked pixel point.

4. The method for identifying the needle tip position of a thyroid fine needle aspiration ultrasound image according to claim 2, wherein: Determining the characteristic state change index corresponding to each pixel point based on the difference between the local difference characteristic value corresponding to each pixel point and the local difference characteristic value corresponding to the corresponding pixel point in the previous frame of the target ultrasound image includes: The absolute value of the difference between the local difference eigenvalue corresponding to each pixel point in each frame of the target ultrasound image and the local difference eigenvalue corresponding to the corresponding pixel point in the previous frame of the target ultrasound image is determined as the initial state change factor corresponding to each pixel point in each frame of the target ultrasound image; According to the initial state change factor and local grayscale deviation factor corresponding to each pixel point in each frame of the target ultrasound image, the characteristic state change index corresponding to each pixel point in each frame of the target ultrasound image is determined, wherein the initial state change factor and the local grayscale deviation factor are both positively correlated with the characteristic state change index.

5. The method for identifying the needle tip position of a thyroid fine needle aspiration ultrasound image according to claim 1, characterized in that: The step of screening out a target change region from each frame of the target ultrasound image according to a characteristic state change index corresponding to a pixel point in each frame of the target ultrasound image comprises: Filtering out pixels whose corresponding characteristic state change index is greater than a preset change threshold from each frame of the target ultrasound image as changed pixels; The minimum circumscribed rectangle of the area formed by all the changed pixels in each frame of the target ultrasound image is determined as the target changed area.

6. The method for identifying the needle tip position of a thyroid fine needle aspiration ultrasound image according to claim 1, characterized in that: The method of screening out the target fine needle region from each frame of the target ultrasound image according to the shape characteristics of all target edge contours in each frame of the target ultrasound image comprises: Determine the local contour direction vector corresponding to each edge pixel point on each target edge contour according to each edge pixel point and its adjacent edge pixel points; The absolute value of the cosine similarity between the local contour direction vectors corresponding to each edge pixel point on each target edge contour and any one of its adjacent edge pixel points is determined as the direction change similarity corresponding to each edge pixel point on each target edge contour; According to the difference between the similarities of the direction changes corresponding to all adjacent edge pixels on each target edge contour, the possible indicators of the fine needle shape corresponding to each target edge contour are determined; Filter out the target edge contour with the largest possible index of the corresponding fine needle shape from each frame of the target ultrasound image as the candidate fine needle contour; The edge pixels on the candidate outline of the fine needle in each frame of the target ultrasound image are used as initial seed points for region growing to obtain the target subregion; The union of the inner area of the thin needle candidate contour and the target sub-area to which the thin needle candidate contour belongs is determined as the target thin needle area.

7. The method for identifying the needle tip position of a thyroid fine needle aspiration ultrasound image according to claim 6, characterized in that: The determining, based on each edge pixel point on each target edge contour and its adjacent edge pixel points, a local contour direction vector corresponding to each edge pixel point on each target edge contour comprises: Determine any target edge contour as a marked edge contour, and determine any edge pixel point on the marked edge contour as a marked edge point; Determine any one edge pixel point among two edge pixel points adjacent to the marked edge point on the marked edge contour as a reference edge point; Based on the marked edge point and the reference edge point, a local contour direction vector corresponding to the marked edge point is constructed, wherein the modulus of the local contour direction vector corresponding to the marked edge point is equal to the distance between the marked edge point and the reference edge point; and the direction of the local contour direction vector corresponding to the marked edge point is the direction from the marked edge point to the reference edge point.

8. The method for identifying the needle tip position of a thyroid fine needle aspiration ultrasound image according to claim 6, characterized in that: Determining the possible indicators of the fine needle shape corresponding to each target edge contour according to the difference between the similarities of the direction changes corresponding to all adjacent edge pixels on each target edge contour includes: The absolute value of the difference between the direction change similarities corresponding to each adjacent edge pixel point on each target edge contour is determined as the change similarity difference, and the change similarity difference set corresponding to each target edge contour is obtained; The average of all the change similarity differences in the change similarity difference set corresponding to each target edge contour is determined as the overall similarity difference corresponding to each target edge contour; According to the overall similarity difference corresponding to each target edge contour, the possible index of the fine needle shape corresponding to each target edge contour is determined, wherein the overall similarity difference is negatively correlated with the possible index of the fine needle shape.

9. The method for identifying the needle tip position of a thyroid fine needle aspiration ultrasound image according to claim 1, characterized in that: The step of determining the fine needle motion state corresponding to each target ultrasound image frame according to the motion change between the target fine needle regions in adjacent target ultrasound images includes: Filter out the pixel point with the largest corresponding characteristic state change index from the target fine needle area in each frame of the target ultrasound image as the needle tip candidate point; Screening out the pixel point with the largest grayscale difference from the candidate needle tip point in a preset neighborhood corresponding to the candidate needle tip point in each frame of the target ultrasound image as the reference point corresponding to the candidate needle tip point in each frame of the target ultrasound image; Constructing a reference direction vector corresponding to each frame of the target ultrasound image based on the candidate needle tip point and its corresponding reference point in each frame of the target ultrasound image, wherein the modulus of the reference direction vector is equal to the distance between the candidate needle tip point and its corresponding reference point, and the direction of the reference direction vector is the direction from the candidate needle tip point to its corresponding reference point; If the cosine similarity between the reference direction vectors corresponding to adjacent target ultrasound images is greater than a preset similarity threshold, it is determined that the fine needle motion states corresponding to the adjacent target ultrasound images are the same; The target ultrasound images of consecutive frames with the same needle motion state are formed into a target ultrasound image group to obtain a target ultrasound image group sequence; setting the fine needle movement states corresponding to all target ultrasound images in the target ultrasound image groups with odd numbers in the target ultrasound image group sequence as fine needle insertion states; The fine needle motion states corresponding to all target ultrasound images in the target ultrasound image groups with even numbers in the target ultrasound image group sequence are set as the fine needle withdrawal state.

10. The method for identifying the needle tip position of a thyroid fine needle aspiration ultrasound image according to claim 9, characterized in that: The identifying of the needle tip position in each frame of the target ultrasound image based on the fine needle motion state corresponding to each frame of the target ultrasound image includes: If the fine needle motion state corresponding to the target ultrasound image is the fine needle insertion state, the position of the needle tip candidate point in the target ultrasound image is determined as the needle tip position in the target ultrasound image; If the fine needle motion state corresponding to the target ultrasound image is the fine needle withdrawal state, the position of the target fine needle region to which the needle tip candidate point belongs in the target ultrasound image is determined as the needle tip position in the target ultrasound image.

Images