An ultrasonic imaging omics analysis method and system for the pathological characteristics of bladder tumors

Abstract

The present invention relates to the technical field of image analysis, and particularly relates to an ultrasonic imaging omics analysis method and system for the pathological characteristics of bladder tumors. The method includes: acquiring a target ultrasonic image corresponding to the bladder of a patient to be detected, and performing connected component segmentation on the target ultrasonic image; determining possible factors of the bladder cavity corresponding to the connected components; determining an abnormal bulge index corresponding to each edge pixel point on the screened bladder cavity connected components; screening out abnormal bulge edges from the bladder cavity connected components, and connecting the two end points of the abnormal bulge edges; determining a bladder tumor characteristic index corresponding to the abnormal bulge edge according to the shape rule of the abnormal bulge edge and the gray scale distribution between the abnormal bulge edge and its corresponding normal edge segment, so as to determine whether each abnormal bulge edge is a bladder tumor edge. By analyzing the pathological characteristics of bladder tumors in the target ultrasonic image, the present invention realizes the recognition of bladder tumors and improves the accuracy of bladder tumor recognition.

Description

Technical Field

[0001] The present invention relates to the technical field of image analysis, and particularly to an ultrasonic imaging omics analysis method and system for pathological features of bladder tumors. Background Art

[0002] Ultrasonic imaging is a non-invasive method that utilizes the principle of high-frequency sound waves penetrating human tissues and reflecting back to generate two-dimensional images of internal tissues and organs in real time. Therefore, the application of ultrasonic imaging is relatively extensive. For example, it can be used to assist doctors in observing the situation of bladder tumors. However, due to the diversity of bladder tumors in terms of morphology, size, location, and growth pattern, it is often difficult for doctors to quickly identify bladder tumors during the observation process. Therefore, in order to facilitate doctors in observing the situation of bladder tumors, it is crucial to assist doctors in identifying bladder tumors based on ultrasonic images. Currently, when identifying an object, the commonly used method is to segment the object region from the object image through the Otsu threshold method.

[0003] However, when segmenting bladder tumors from ultrasonic images through the Otsu threshold method, the following technical problems often exist:

[0004] Since the gray-scale difference between the bladder tumor region and the normal bladder region is often small, and the gray-scale of foreign objects such as stones and polyps in ultrasonic images is often similar to that of the bladder tumor region, when using the Otsu threshold method to segment bladder tumors based on the difference in gray-scale values, misjudgment of bladder tumor pixel points may occur, resulting in difficulty in accurately segmenting the bladder tumor region and poor accuracy in identifying bladder tumors. Summary of the Invention

[0005] In order to solve the technical problem of poor accuracy in identifying bladder tumors, the present invention proposes an ultrasonic imaging omics analysis method and system for pathological features of bladder tumors.

[0006] In a first aspect, the present invention provides an ultrasonic imaging omics analysis method for pathological features of bladder tumors, the method comprising:

[0007] Obtaining a target ultrasonic image corresponding to the bladder of a patient to be detected, and performing connected component division on the target ultrasonic image;

[0008] Determining a possible bladder cavity factor corresponding to each connected component according to the position of each connected component and the gray-scale distribution of each connected component, and screening out the bladder cavity connected components from all connected components based on the possible bladder cavity factor corresponding to the connected component;

[0009] Determine the abnormal bulge index corresponding to each edge pixel point on the bladder cavity connected domain according to the gray - level difference between the inner and outer sides of each edge pixel point on the bladder cavity connected domain and the edge tangent slope difference within the preset neighborhood edge segment corresponding to each edge pixel point on the bladder cavity connected domain;

[0010] Based on the abnormal bulge indexes corresponding to all edge pixel points on the bladder cavity connected domain, screen out the abnormal bulge edges from the bladder cavity connected domain, and connect the two endpoints of each abnormal bulge edge to obtain the normal edge line segment corresponding to each abnormal bulge edge;

[0011] Determine the bladder tumor feature index corresponding to each abnormal bulge edge according to the shape - rule condition of each abnormal bulge edge and the gray - level distribution between each abnormal bulge edge and its corresponding normal edge line segment;

[0012] Judge whether each abnormal bulge edge is a bladder tumor edge according to the bladder tumor feature index corresponding to each abnormal bulge edge.

[0013] Combined with the above - mentioned first aspect, in a possible implementation manner, the determining the possible bladder cavity factor corresponding to each connected domain according to the position of each connected domain and the gray - level distribution of each connected domain includes:

[0014] Determine any connected domain in the target ultrasonic image as the marked connected domain, and determine the center point of the target ultrasonic image as the reference center point;

[0015] Determine the gray - level representative value corresponding to the marked connected domain as the gray - level value with the largest number of pixel points in the marked connected domain;

[0016] Determine the possible bladder cavity factor corresponding to the marked connected domain according to the distance between the center point of the marked connected domain and the reference center point, the gray - level representative value corresponding to the marked connected domain, and the number of pixel points in the marked connected domain whose corresponding gray - level values are equal to its corresponding gray - level representative value, where the center point of the marked connected domain is used to represent the position of the marked connected domain.

[0017] Combined with the above - mentioned first aspect, in a possible implementation manner, the screening out the bladder cavity connected domain from all connected domains based on the possible bladder cavity factor corresponding to the connected domain includes:

[0018] Screen out the connected domain with the largest possible bladder cavity factor corresponding to it from all connected domains as the bladder cavity connected domain.

[0019] Combined with the above - mentioned first aspect, in a possible implementation manner, the formula for the possible bladder cavity factor corresponding to the marked connected domain is:

[0020] ;

[0021] Among them, is a possible factor of the bladder cavity corresponding to the labeled connected region; is the number of pixel points whose corresponding gray values in the labeled connected region are equal to their corresponding gray representative values; is an exponential function with the natural constant as the base; is the gray representative value corresponding to the labeled connected region; is the distance between the center point of the labeled connected region and the reference center point.

[0022] Combined with the above first aspect, in a possible implementation manner, determining the abnormal protrusion index corresponding to each edge pixel point on the bladder cavity connected region according to the gray difference between the inner and outer sides of each edge pixel point on the bladder cavity connected region, and the edge tangent slope difference within the preset neighborhood edge segment corresponding to each edge pixel point on the bladder cavity connected region includes:

[0023] Determine any edge pixel point on the bladder cavity connected region as a marked edge point, and select a preset number of pixel points closest to the marked edge point on both sides of the marked edge point from the normal line of the marked edge point to form two pixel point sequences corresponding to the marked edge point;

[0024] Determine the gray difference between the inner and outer sides corresponding to the marked edge point according to the gray difference between the two pixel point sequences corresponding to the marked edge point;

[0025] Determine the edge tangent slope difference corresponding to the marked edge point according to the difference between the slopes of the tangents corresponding to adjacent edge pixel points within the preset neighborhood edge segment corresponding to the marked edge point;

[0026] Determine the abnormal protrusion index corresponding to the marked edge point according to the gray difference between the inner and outer sides and the edge tangent slope difference corresponding to the marked edge point.

[0027] Combined with the above first aspect, in a possible implementation manner, the formula corresponding to the abnormal protrusion index of the marked edge point is:

[0028] ;

[0029] ;

[0030] ;

[0031] Among them, is the abnormal protrusion index corresponding to the marked edge point; is the absolute value function; is the gray difference between the inner and outer sides corresponding to the marked edge point; is the mean value of the gray - scale differences between the inner and outer sides corresponding to all edge pixel points on the bladder cavity connected region; is the difference in the edge tangent slopes corresponding to the marked edge points; is the mean value of the differences in the edge tangent slopes corresponding to all edge pixel points on the bladder cavity connected region; is the mean value of the gray - scale values corresponding to all pixel points in any one pixel point sequence corresponding to the marked edge points; is the mean value of the gray - scale values corresponding to all pixel points in another pixel point sequence corresponding to the marked edge points; is the number of edge pixel points within the preset neighborhood edge segment; is the serial number of the edge pixel points within the preset neighborhood edge segment; is within the preset neighborhood edge segment corresponding to the marked edge points, the slope of the tangent line corresponding to the th edge pixel point; is the slope of the tangent line corresponding to the

[0032] Combined with the above - mentioned first aspect, in a possible implementation manner, the screening of the abnormal bulge edges from the bladder cavity connected region based on the abnormal bulge indexes corresponding to all edge pixel points on the bladder cavity connected region includes:

[0033] If the abnormal bulge index corresponding to the edge pixel point is greater than the preset abnormal threshold, then the edge pixel point is determined as an abnormal edge point;

[0034] The continuous abnormal edge points are formed into an abnormal bulge edge.

[0035] Combined with the above - mentioned first aspect, in a possible implementation manner, the determination of the bladder tumor characteristic index corresponding to each abnormal bulge edge according to the shape - rule condition of each abnormal bulge edge and the gray - scale distribution between each abnormal bulge edge and its corresponding normal edge segment includes:

[0036] Any one abnormal bulge edge is determined as a marked bulge edge, and the edge chain - code sequence corresponding to the marked bulge edge is obtained, where the edge chain - code sequence corresponding to the marked bulge edge is used to characterize the shape - rule condition of the marked bulge edge;

[0037] The normal line of the center point of the marked bulge edge is determined as the target normal line corresponding to the marked bulge edge, and the intersection point between the target normal line and the normal edge segment corresponding to the marked bulge edge is determined as the target intersection point corresponding to the marked bulge edge;

[0038] Connect the center point of the edge of the marked protrusion to the target intersection point to obtain the target line segment corresponding to the edge of the marked protrusion, where the starting point of the target line segment is the center point of the edge of the marked protrusion, and the ending point of the target line segment is the target intersection point;

[0039] Determine the bladder tumor characteristic index corresponding to the edge of the marked protrusion according to the edge chain code sequence corresponding to the edge of the marked protrusion and the gray-scale distribution on the target line segment corresponding to the edge of the marked protrusion.

[0040] Combined with the above first aspect, in a possible implementation manner, the formula corresponding to the bladder tumor characteristic index corresponding to the edge of the marked protrusion is:

[0041] ;

[0042] where is the bladder tumor characteristic index corresponding to the edge of the marked protrusion; is the normalization function; is the mean value of all chain codes in the edge chain code sequence corresponding to the edge of the marked protrusion; is the number of pixel points on the target line segment corresponding to the edge of the marked protrusion; is the serial number of the pixel point on the target line segment corresponding to the edge of the marked protrusion; is the th pixel point on the target line segment corresponding to the edge of the marked protrusion; is the th pixel point on the target line segment corresponding to the edge of the marked protrusion; is a preset factor greater than 0.

[0043] In a second aspect, the present invention provides an ultrasound imaging omics analysis system for bladder tumor pathological features, and the system includes:

[0044] An acquisition and partitioning module, configured to acquire a target ultrasound image corresponding to the bladder of a patient to be detected, and perform connected component partitioning on the target ultrasound image;

[0045] A determination and screening module, configured to determine the possible bladder cavity factor corresponding to each connected component according to the position of each connected component and the gray-scale distribution of each connected component, and screen out the bladder cavity connected components from all connected components based on the possible bladder cavity factor corresponding to the connected component;

[0046] An abnormal protrusion index determination module, configured to determine the abnormal protrusion index corresponding to each edge pixel point on the bladder cavity connected component according to the gray-scale difference between the inner and outer sides of each edge pixel point on the bladder cavity connected component and the edge tangent slope difference within the preset neighborhood edge segment corresponding to each edge pixel point on the bladder cavity connected component;

[0047] A screening and connection module, configured to screen out abnormal bulge edges from the bladder cavity connected region based on the abnormal bulge indexes corresponding to all edge pixels on the bladder cavity connected region, and connect two end points of each abnormal bulge edge to obtain a normal edge line segment corresponding to each abnormal bulge edge;

[0048] A bladder tumor feature index determination module, configured to determine a bladder tumor feature index corresponding to each abnormal bulge edge according to the shape regularity of each abnormal bulge edge and the gray-scale distribution between each abnormal bulge edge and its corresponding normal edge line segment;

[0049] A bladder tumor edge judgment module, configured to judge whether each abnormal bulge edge is a bladder tumor edge according to the bladder tumor feature index corresponding to each abnormal bulge edge.

[0050] In a third aspect, a server is provided, including a memory and a processor. The memory is used to store executable program codes, and the processor is used to call and run the executable program codes from the memory, so that the device executes the method in the above first aspect or any possible implementation manner of the first aspect.

[0051] In a fourth aspect, a computer program product is provided, including: computer program codes, when the computer program codes are run on a computer, enabling the computer to execute the method in the above first aspect or any possible implementation manner of the first aspect.

[0052] In a fifth aspect, a computer-readable storage medium is provided, storing computer program codes, when the computer program codes are run on a computer, enabling the computer to execute the method in the above first aspect or any possible implementation manner of the first aspect.

[0053] The present invention has the following beneficial effects:

[0054] An ultrasonic imaging genomics analysis method for pathological features of bladder tumors according to the present invention realizes the recognition of bladder tumors by analyzing the pathological features of bladder tumors in a target ultrasonic image, solves the technical problem of poor accuracy in bladder tumor recognition, and improves the accuracy of bladder tumor recognition. Compared with using the Otsu threshold method to segment bladder tumors using different gray-scale values, the present invention comprehensively considers multiple factors related to bladder tumor features, such as gray-scale difference, edge tangent slope difference, abnormal bulge index, shape regularity, and bladder tumor feature index, etc., thereby improving the accuracy of bladder tumor recognition and better assisting doctors in observing bladder tumors. Description of the Drawings

[0055] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0056] Figure 1 It is a flowchart of an ultrasonic imaging omics analysis method for the pathological characteristics of bladder tumors according to the present invention;

[0057] Figure 2 It is a schematic diagram of the composition structure of an ultrasonic imaging omics analysis system for the pathological characteristics of bladder tumors according to the present invention;

[0058] Figure 3 It is a schematic diagram of the structure of a computer device according to the present invention. Detailed implementation manners

[0059] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following, in combination with the drawings and preferred embodiments, details the specific implementation manners, structures, features and their effects of the technical solutions proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0060] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs.

[0061] Reference Figure 1 , which shows the flow of some embodiments of an ultrasonic imaging omics analysis method for the pathological characteristics of bladder tumors according to the present invention. The ultrasonic imaging omics analysis method for the pathological characteristics of bladder tumors includes the following steps:

[0062] Step S1, obtain the target ultrasonic image corresponding to the bladder of the patient to be detected, and perform connected component division on the target ultrasonic image.

[0063] Among them, the patient to be detected can be a patient to be subjected to bladder tumor detection. The target ultrasonic image can be the ultrasonic image of the bladder of the patient to be detected after image preprocessing. Image preprocessing can include, but is not limited to, denoising processing and image enhancement.

[0064] It should be noted that when the patient's bladder is in a filled state, a high-frequency ultrasound device is used to scan the patient's bladder to obtain more detailed ultrasound images of the patient's bladder. The normal bladder wall usually appears as a bright echo band in the ultrasound image, and when the bladder is filled, the urine in the cavity is often anechoic; however, bladder tumors usually appear as abnormal echo masses protruding into the bladder cavity in the ultrasound image. The ultrasound device performs ultrasound imaging on the patient's bladder based on the echo performance. Image preprocessing of the image data can improve the image quality.

[0065] As an example, this step may include the following steps:

[0066] First, through the ultrasound device, collect the ultrasound image of the bladder of the patient to be detected, perform denoising processing on the ultrasound image, perform image enhancement on the denoised ultrasound image, and use the ultrasound image after image enhancement at this time as the target ultrasound image.

[0067] Second, through the connected component extraction algorithm, perform connected component extraction on the target ultrasound image to obtain multiple connected components, realizing the division of connected components.

[0068] Step S2, according to the position of each connected component and the gray-scale distribution of each connected component, determine the possible factor of the bladder cavity corresponding to each connected component, and based on the possible factor of the bladder cavity corresponding to the connected component, screen out the connected component of the bladder cavity from all connected components.

[0069] Among them, the connected component of the bladder cavity can be used to represent the bladder cavity.

[0070] It should be noted that in the patient's bladder wall, tumors often grow inside the bladder cavity and are closely attached to the bladder wall, which appears as a structure protruding into the bladder cavity in the ultrasound image, resulting in local thickening of the bladder wall and possible abnormal connection with the surrounding tissues. Therefore, identifying the bladder cavity where bladder tumors may exist can reduce the interference of irrelevant information to a certain extent, making the tumor characteristics more prominent, so as to facilitate the subsequent accurate identification of bladder tumors.

[0071] As an example, this step may include the following steps:

[0072] First, determine any connected component in the above target ultrasound image as the marked connected component, and determine the center point of the above target ultrasound image as the reference center point.

[0073] Second, determine the gray-scale representative value corresponding to the above marked connected component as the gray-scale value with the most pixels in the above marked connected component.

[0074] Step 3: Determine the possible bladder cavity factor corresponding to the marked connected region according to the distance between the center point of the marked connected region and the reference center point, the grayscale representative value corresponding to the marked connected region, and the number of pixel points in the marked connected region whose corresponding grayscale value is equal to its corresponding grayscale representative value.

[0075] Among them, the center point of the marked connected region can be used to characterize the position of the marked connected region.

[0076] It should be noted that in ultrasonic images, the bladder usually appears as a liquid anechoic area, which is often surrounded by a hyperechoic bladder wall. Bladder tumors usually grow attached to the inner wall of the bladder and grow towards the cavity, appearing as a protrusion in ultrasonic images. The bladder cavity often presents as a connected region with a relatively dark color and a relatively large area. Secondly, the human bladder usually has a fixed and similar shape. The extravesical tissue and the part of the bladder tumor often present the characteristic of a light color. First, perform connected region segmentation on the ultrasonic image of the patient's bladder. In the ultrasonic image, the liquid in the bladder cavity, the extravesical tissue, and the foreign objects in the cavity should be divided into different connected regions in order to segment the target abnormal region and facilitate the doctor's diagnosis. However, it is found in actual observation that the connected region segmentation result cannot fully achieve accurate segmentation. When the patient has a bladder tumor, it may be connected to the extravesical tissue to form the same connected region. That is to say, the connected region of the bladder cavity divided often contains the bladder cavity and the foreign objects in the bladder cavity. The foreign objects in the bladder cavity can include bladder tumors, stones, polyps, etc. Therefore, in the subsequent embodiments of the present invention, the bladder tumor region is mainly segmented adaptively by analyzing the connected region of the bladder cavity.

[0077] For example, the formula for determining the possible bladder cavity factor corresponding to the marked connected region can be:

[0078] ;

[0079] Among them, is the possible bladder cavity factor corresponding to the marked connected region. is the number of pixel points in the marked connected region whose corresponding grayscale value is equal to its corresponding grayscale representative value, that is, the number of pixel points in the marked connected region whose corresponding grayscale value is equal to . is the exponential function with the natural constant as the base. is the grayscale representative value corresponding to the marked connected region. is the distance between the center point of the marked connected region and the reference center point.

[0080] It should be noted that in actual situations, the bladder cavity often presents as a connected region with a relatively dark color and a relatively large area. Therefore, the gray levels of the pixel points in the bladder cavity tend to be close to zero. Moreover, when the doctor moves the ultrasound probe on the patient's body surface with the bladder as the target, the bladder is always close to the center position of the ultrasound image. When is larger, it often indicates that there are more pixel points with the corresponding gray level value equal to in the marked connected region. When is smaller, it often indicates that the overall gray level of the marked connected region is smaller. When is smaller, it often indicates that the center point of the marked connected region is closer to the center position of the ultrasound image. Therefore, when is larger, it often indicates that the overall color of the marked connected region is darker and the area of its darker-colored region is larger, and it is closer to the center position of the ultrasound image; it often indicates that the marked connected region is more likely to represent the bladder cavity.

[0081] Step 4: Select the connected region with the largest corresponding bladder cavity possible factor from all the connected regions as the bladder cavity connected region.

[0082] Step S3: Determine the abnormal protrusion index corresponding to each edge pixel point on the bladder cavity connected region according to the gray level difference between the inner and outer sides of each edge pixel point on the bladder cavity connected region and the edge tangent slope difference within the preset neighborhood edge segment corresponding to each edge pixel point on the bladder cavity connected region.

[0083] Among them, the preset neighborhood edge segment corresponding to the edge pixel point can be a sub-edge intercepted from the edge of the bladder cavity connected region and located around the edge pixel point. The edge pixel point can be located at the center of its corresponding preset neighborhood edge segment. The number of edge pixel points in the preset neighborhood edge segment can be a preset number. For example, the number of edge pixel points in the preset neighborhood edge segment can be 15. The number of edge pixel points in the preset neighborhood edge segment can also be equal to the average length of the bladder tumors detected historically.

[0084] It should be noted that bladder tumors usually grow inside the bladder cavity and cling to the bladder wall. In ultrasonic images, they appear as structures protruding into the bladder cavity, causing local thickening of the bladder wall and possible abnormal connections with surrounding tissues. The bladder wall at the location of the bladder tumor often shows significant bulging changes. Such morphological changes result in a greater difference in the tangent slope of the pixel points at the tumor edge compared to the surrounding healthy tissues. That is to say, the change in the tangent slope of the pixel points at the tumor edge is often more drastic, showing a relatively large difference from the overall performance. In addition, since bladder tumors are diseased tissues, there are often certain differences in the color distribution shown in ultrasonic images between them and the healthy bladder wall. That is to say, when the gray-scale change of the pixel points on both sides of the bladder wall is abnormal, it often indicates that there may be a bladder tumor or other foreign objects at that location of the bladder wall. Therefore, the larger the abnormal bulging index corresponding to the quantified edge pixel points, the more likely it is that the edge pixel points are the edge pixel points of the bladder tumor.

[0085] As an example, this step may include the following steps:

[0086] First step, determine any edge pixel point on the connected domain of the bladder cavity as a marked edge point. From the normal line of the above-mentioned marked edge point and on both sides of the marked edge point, screen out the preset number of pixel points closest to the marked edge point respectively to form two pixel point sequences corresponding to the above-mentioned marked edge point.

[0087] Among them, the preset number can be the number of pixel points set in advance. For example, the preset number can be 10. A pixel point sequence can be composed of the preset number of pixel points on one side of the marked edge point.

[0088] Second step, determine the gray-scale difference between the inside and outside corresponding to the above-mentioned marked edge point according to the gray-scale difference between the two pixel point sequences corresponding to the above-mentioned marked edge point.

[0089] Third step, determine the edge tangent slope difference corresponding to the above-mentioned marked edge point according to the difference between the slopes of the tangents corresponding to adjacent edge pixel points within the preset neighborhood edge segment corresponding to the above-mentioned marked edge point.

[0090] Fourth step, determine the abnormal bulging index corresponding to the above-mentioned marked edge point according to the gray-scale difference between the inside and outside and the edge tangent slope difference corresponding to the above-mentioned marked edge point.

[0091] For example, the formula for determining the abnormal bulging index corresponding to the marked edge point can be:

[0092] ;

[0093] ;

[0094] ;

[0095] Among them, is the abnormal protrusion index corresponding to the marked edge point. is the absolute value function. is the gray level difference between the inner and outer sides corresponding to the marked edge point. is the average value of the gray level differences between the inner and outer sides corresponding to all edge pixel points on the bladder cavity connected region. is the difference in the edge tangent slope corresponding to the marked edge point. is the average value of the differences in the edge tangent slopes corresponding to all edge pixel points on the bladder cavity connected region. is the average value of the gray level values corresponding to all pixel points in any pixel point sequence corresponding to the marked edge point. is the average value of the gray level values corresponding to all pixel points in another pixel point sequence corresponding to the marked edge point. is the number of edge pixel points within the preset neighborhood edge segment. is the serial number of the edge pixel points within the preset neighborhood edge segment. is within the preset neighborhood edge segment corresponding to the marked edge point, the th is the slope corresponding to the tangent of the th edge pixel point.

[0096] It should be noted that can characterize the gray level change of the pixel points on both sides of the bladder wall where the marked edge point is located. can characterize the overall gray level change of the pixel points on both sides of the bladder wall. Therefore, when is larger, it often indicates that the difference between the gray level change of the pixel points on both sides of the bladder wall where the marked edge point is located and the overall gray level change of the pixel points on both sides of the bladder wall is larger; it often indicates that the marked edge point is more likely to be an abnormal protrusion edge pixel point, and it often indicates that the marked edge point is more likely to be a tumor edge pixel point. can characterize the edge slope change situation around the marked edge point. can characterize the overall edge slope change situation of the bladder wall. Therefore, when is larger, it often indicates that the difference between the edge slope change degree around the marked edge point and the overall edge slope change degree of the bladder wall is larger; it often indicates that the marked edge point is more likely to be an abnormal protrusion edge pixel point, and it often indicates that the marked edge point is more likely to be a tumor edge pixel point. Therefore, when is larger, it often indicates that the marked edge point is more likely to be an abnormal protrusion edge pixel point, and it often indicates that the marked edge point is more likely to be a tumor edge pixel point.

[0097] Step S4: Based on the abnormal bulge indices corresponding to all the edge pixels of the bladder cavity connected domain, filter out the abnormal bulge edges from the bladder cavity connected domain, and connect the two endpoints of each abnormal bulge edge to obtain the normal edge segments corresponding to each abnormal bulge edge.

[0098] Among them, the closed area formed by an abnormal bulge edge and its corresponding normal edge segment can represent a foreign object. The abnormal bulge edge can be the edge bulging on the bladder wall. The normal edge segment can represent the bladder wall edge blocked by the foreign object.

[0099] As an example, this step may include the following steps:

[0100] First step: If the abnormal bulge index corresponding to the edge pixel is greater than the preset abnormal threshold, then determine the edge pixel as an abnormal edge point.

[0101] Among them, the preset abnormal threshold can be a threshold set in advance for judging whether a pixel is abnormal. For example, the preset abnormal threshold can be 0.6.

[0102] Second step: Form the abnormal bulge edges with consecutive abnormal edge points.

[0103] Third step: Connect the two endpoints of each abnormal bulge edge to obtain the normal edge segments corresponding to each abnormal bulge edge.

[0104] Step S5: According to the shape rule situation of each abnormal bulge edge and the gray-scale distribution between each abnormal bulge edge and its corresponding normal edge segment, determine the bladder tumor characteristic index corresponding to each abnormal bulge edge.

[0105] It should be noted that in addition to bladder tumors, there may be foreign objects such as stones and polyps in the bladder cavity. These foreign objects may form strong echo masses or medium echo areas on the ultrasound image, and their positions in the bladder cavity are not fixed. When they are close to the bladder wall, they often interfere with the recognition of the changes in the bladder wall edge caused by bladder tumors. Therefore, for the connected domains corresponding to all the bulge edges, that is, the areas formed by the abnormal bulge edges and their corresponding normal edge segments, when performing bladder tumor recognition on these areas, bladder tumor characteristic analysis is often required. Generally speaking, the edges of bladder tumors are often irregular, and when collecting ultrasound images, the patient's bladder is often in a filled state. In this state, due to its growth characteristics, the bladder tumor will infiltrate into the bladder wall to a greater extent, and this infiltration is manifested as a gradually decreasing trend in the gray-scale value between the bulge edge of the bladder tumor and its corresponding bladder wall on the ultrasound image.

[0106] As an example, this step may include the following steps:

[0107] First, determine any abnormal convex edge as the marked convex edge, and obtain the edge chain code sequence corresponding to the above-mentioned marked convex edge.

[0108] Among them, the edge chain code sequence corresponding to the above-mentioned marked convex edge can be used to characterize the shape regularity of the above-mentioned marked convex edge. The edge chain code sequence can be obtained through the Freeman chain code algorithm.

[0109] Second, determine the normal line of the center point of the above-mentioned marked convex edge as the target normal line corresponding to the above-mentioned marked convex edge, and determine the intersection point between the above-mentioned target normal line and the normal edge segment corresponding to the above-mentioned marked convex edge as the target intersection point corresponding to the above-mentioned marked convex edge.

[0110] Third, connect the center point of the above-mentioned marked convex edge and the above-mentioned target intersection point to obtain the target line segment corresponding to the above-mentioned marked convex edge.

[0111] Among them, the starting point of the target line segment can be the center point of the above-mentioned marked convex edge, and the ending point of the target line segment can be the above-mentioned target intersection point.

[0112] Fourth, determine the bladder tumor characteristic index corresponding to the above-mentioned marked convex edge according to the edge chain code sequence corresponding to the above-mentioned marked convex edge and the gray-scale distribution on the target line segment corresponding to the above-mentioned marked convex edge.

[0113] For example, the formula for determining the bladder tumor characteristic index corresponding to the marked convex edge can be:

[0114] ;

[0115] Among them, is the bladder tumor characteristic index corresponding to the marked convex edge. is the normalization function. is the mean value of all chain codes in the edge chain code sequence corresponding to the marked convex edge. is the number of pixel points on the target line segment corresponding to the marked convex edge. is the serial number of the pixel point on the target line segment corresponding to the marked convex edge. is the th pixel point corresponding gray value on the target line segment corresponding to the marked convex edge. is the th pixel point corresponding gray value on the target line segment corresponding to the marked convex edge. is a preset factor greater than 0, mainly used to prevent the denominator from being 0. For example, can be 0.001.

[0116] It should be noted that when The larger it is, it often indicates that the chain code in the edge chain code sequence corresponding to the marked convex edge is larger, which often indicates that the marked convex edge is more irregular, and often indicates that the marked convex edge is more likely to be the edge of a bladder tumor. When The larger it is, it often indicates that the gray values corresponding to the pixel points on the target line segment corresponding to the marked convex edge show a decreasing trend. Therefore, when The larger it is, it often indicates that the degree of decrease in the gray values corresponding to the pixel points on the target line segment corresponding to the marked convex edge is greater, which often indicates that the gray values between the marked convex edge and its corresponding normal edge line segment show a gradually decreasing situation, and often indicates that the marked convex edge is more likely to be the edge of a bladder tumor. Therefore, when The larger it is, it often indicates that the marked convex edge is more likely to be the edge of a bladder tumor.

[0117] Step S6: According to the bladder tumor characteristic indexes corresponding to each abnormal convex edge, determine whether each abnormal convex edge is the edge of a bladder tumor.

[0118] As an example, if the bladder tumor characteristic index corresponding to the abnormal convex edge is greater than the preset tumor characteristic threshold, then the abnormal convex edge is determined to be the edge of a bladder tumor. Among them, the preset tumor characteristic threshold can be a threshold set in advance for tumor judgment. For example, the preset tumor characteristic threshold can be 0.8.

[0119] Optionally, for the convenience of doctors' observation, different colors can be used to mark the abnormal convex edges corresponding to different bladder tumor characteristic indexes and their corresponding normal edge line segments, making the pathological features of the patient clearer and more intuitive, and providing auxiliary diagnostic information on the pathological features of bladder tumors for doctors. This method not only enhances the readability of the image, but also enables doctors to quickly capture key information, and this information helps doctors formulate personalized treatment plans, improve the treatment effect and the quality of life of patients.

[0120] It should be noted that due to the diversity of bladder tumors in terms of morphology, size, location and growth pattern, doctors face great challenges in the diagnosis process. Therefore, accurately identifying tumors and their boundaries in ultrasound images is crucial. The subtle feature manifestations of tumors and the detailed delineation of tumor boundaries in ultrasound images are often the prerequisites for revealing key information such as the morphology, size, location of tumors and their relationship with surrounding tissues. These information are important references for doctors to evaluate tumor pathological features, formulate treatment plans and predict the prognosis of patients. Therefore, achieving accurate identification and boundary delineation of bladder tumors in ultrasound images is the key to improving the diagnostic accuracy and treatment effect of bladder tumors.

[0121] Reference Figure 2, based on the same inventive concept as the above method embodiments, the present invention provides an ultrasound radiomics analysis system for bladder tumor pathological features. The system includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the computer program is executed by the processor, it implements the steps of an ultrasound radiomics analysis method for bladder tumor pathological features, which may specifically include:

[0122] An acquisition and division module 201, configured to acquire a target ultrasound image corresponding to the bladder of a patient to be detected, and perform connected component division on the target ultrasound image;

[0123] A determination and screening module 202, configured to determine a possible bladder cavity factor corresponding to each connected component according to the position of each connected component and the gray-scale distribution of each connected component, and screen out the bladder cavity connected components from all connected components based on the possible bladder cavity factor corresponding to the connected component;

[0124] An abnormal protrusion index determination module 203, configured to determine an abnormal protrusion index corresponding to each edge pixel point on the bladder cavity connected component according to the gray-scale difference between the inner and outer sides of each edge pixel point on the bladder cavity connected component and the edge tangent slope difference within a preset neighborhood edge segment corresponding to each edge pixel point on the bladder cavity connected component;

[0125] A screening and connection module 204, configured to screen out abnormal protrusion edges from the bladder cavity connected component based on the abnormal protrusion indexes corresponding to all edge pixel points on the bladder cavity connected component, and connect the two end points of each abnormal protrusion edge to obtain a normal edge line segment corresponding to each abnormal protrusion edge;

[0126] A bladder tumor feature index determination module 205, configured to determine a bladder tumor feature index corresponding to each abnormal protrusion edge according to the shape regularity of each abnormal protrusion edge and the gray-scale distribution between each abnormal protrusion edge and its corresponding normal edge line segment;

[0127] A bladder tumor edge judgment module 206, configured to judge whether each abnormal protrusion edge is a bladder tumor edge according to the bladder tumor feature index corresponding to each abnormal protrusion edge.

[0128] Figure 3 It is a schematic structural diagram of a computer device provided by an embodiment of the present invention. Exemplarily, as Figure 3 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. When the processor 302 executes the computer program 303, the computer device can execute any of the above-introduced ultrasound radiomics analysis methods for bladder tumor pathological features.

[0129] Based on the same inventive concept as the above method embodiments, the present invention provides a server, including a memory and a processor. The memory is used to store executable program code, and the processor is used to call and run the executable program code from the memory, so that the device executes any one of the above ultrasonic imaging omics analysis methods for the pathological characteristics of bladder tumors.

[0130] Based on the same inventive concept as the above method embodiments, the present invention provides a computer program product, which includes: computer program code. When the computer program code runs on a computer, it causes the computer to execute any one of the above ultrasonic imaging omics analysis methods for the pathological characteristics of bladder tumors.

[0131] Based on the same inventive concept as the above method embodiments, the present invention provides a computer-readable storage medium, which stores computer program code. When the computer program code runs on a computer, it causes the computer to execute any one of the above ultrasonic imaging omics analysis methods for the pathological characteristics of bladder tumors.

[0132] In summary, compared with using the Otsu threshold method to segment bladder tumors based on different gray values, the present invention comprehensively considers multiple factors related to the characteristics of bladder tumors, such as gray difference, edge tangent slope difference, abnormal bulge index, shape regularity, and bladder tumor characteristic index, etc., thereby improving the accuracy of bladder tumor recognition and better assisting doctors in observing bladder tumors.

[0133] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the various embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. An ultrasonic imaging omics analysis method for the pathological characteristics of bladder tumors, characterized in that Including the following steps: Obtain a target ultrasound image corresponding to the bladder of the patient to be detected, and perform connected component division on the target ultrasound image; Determine the possible bladder cavity factor corresponding to each connected component according to the position of each connected component and the gray-scale distribution of each connected component, and screen out the bladder cavity connected components from all connected components based on the possible bladder cavity factor corresponding to the connected component; Determine the abnormal bulge index corresponding to each edge pixel point on the bladder cavity connected component according to the gray-scale difference between the inner and outer sides of each edge pixel point on the bladder cavity connected component and the edge tangent slope difference within the preset neighborhood edge segment corresponding to each edge pixel point on the bladder cavity connected component; Based on the abnormal bulge indexes corresponding to all edge pixel points on the bladder cavity connected component, screen out the abnormal bulge edges from the bladder cavity connected component, and connect the two end points of each abnormal bulge edge to obtain the normal edge line segment corresponding to each abnormal bulge edge; Determine the bladder tumor characteristic index corresponding to each abnormal bulge edge according to the shape rule condition of each abnormal bulge edge and the gray-scale distribution between each abnormal bulge edge and its corresponding normal edge line segment; Judge whether each abnormal bulge edge is a bladder tumor edge according to the bladder tumor characteristic index corresponding to each abnormal bulge edge; The determining the possible bladder cavity factor corresponding to each connected component according to the position of each connected component and the gray-scale distribution of each connected component includes: Determine any connected component in the target ultrasound image as the marked connected component, and determine the center point of the target ultrasound image as the reference center point; Determine the gray-scale representative value corresponding to the marked connected component as the gray-scale value with the largest number of pixel points in the marked connected component; Determine the possible bladder cavity factor corresponding to the marked connected component according to the distance between the center point of the marked connected component and the reference center point, the gray-scale representative value corresponding to the marked connected component, and the number of pixel points in the marked connected component whose corresponding gray-scale value is equal to its corresponding gray-scale representative value, wherein the center point of the marked connected component is used to represent the position of the marked connected component.

2. The ultrasonic imaging omics analysis method for the pathological features of bladder tumors according to claim 1, wherein The screening out the bladder cavity connected components from all connected components based on the possible bladder cavity factor corresponding to the connected component includes: Screen out the connected component with the largest possible bladder cavity factor corresponding to it from all connected components as the bladder cavity connected component.

3. The ultrasound imaging omics analysis method for the pathological characteristics of bladder tumors according to claim 1, characterized in that, The formula for the possible bladder cavity factor corresponding to the marked connected component is: ； Among them, is the possible factor of the bladder cavity corresponding to the labeled connected component; is the number of pixel points in the labeled connected component whose corresponding gray value is equal to its corresponding gray representative value; is the exponential function with the natural constant as the base; is the gray representative value corresponding to the labeled connected component; is the distance between the center point of the labeled connected component and the reference center point.

4. The ultrasonic imaging omics analysis method for pathological features of bladder tumors according to claim 1, characterized in that The determining the abnormal bulge index corresponding to each edge pixel point on the bladder cavity connected component according to the gray-scale difference between the inner and outer sides of each edge pixel point on the bladder cavity connected component and the edge tangent slope difference within the preset neighborhood edge segment corresponding to each edge pixel point on the bladder cavity connected component includes: Determine any edge pixel point on the bladder cavity connected component as the marked edge point, and screen out the preset number of pixel points closest to the marked edge point on both sides of the marked edge point from the normal line of the marked edge point to form two pixel point sequences corresponding to the marked edge point; Determine the gray-scale difference between the inner and outer sides corresponding to the marked edge point according to the gray-scale difference between the two pixel point sequences corresponding to the marked edge point; Determine the edge tangent slope difference corresponding to the marked edge point according to the difference between the slopes corresponding to the tangents of adjacent edge pixels within the preset neighborhood edge segment corresponding to the marked edge point; Determine the abnormal protrusion index corresponding to the marked edge point according to the gray-scale difference between the inner and outer sides corresponding to the marked edge point and the edge tangent slope difference; 5. The ultrasonic imaging omics analysis method for the pathological features of bladder tumors according to claim 4, characterized in that, The formula corresponding to the abnormal protrusion index corresponding to the marked edge point is: ； ； ； Among them, is the abnormal protrusion index corresponding to the marked edge point; is the absolute value function; is the gray-scale difference between the inner and outer sides corresponding to the marked edge point; is the average value of the gray-scale differences between the inner and outer sides corresponding to all edge pixel points on the bladder cavity connected region; is the edge tangent slope difference corresponding to the marked edge point; is the average value of the edge tangent slope differences corresponding to all edge pixel points on the bladder cavity connected region; is the average value of the gray-scale values corresponding to all pixel points in any pixel point sequence corresponding to the marked edge point; is the average value of the gray-scale values corresponding to all pixel points in another pixel point sequence corresponding to the marked edge point; is the number of edge pixel points within the preset neighborhood edge segment; is the serial number of the edge pixel points within the preset neighborhood edge segment; is within the preset neighborhood edge segment corresponding to the marked edge point, the th slope of the tangent line corresponding to the edge pixel point; is within the preset neighborhood edge segment corresponding to the marked edge point, the th slope of the tangent line corresponding to the edge pixel point.

6. The ultrasound imaging omics analysis method for the pathological characteristics of bladder tumors according to claim 1, wherein The screening of the abnormal protrusion edge from the bladder cavity connected region based on the abnormal protrusion indexes corresponding to all edge pixels on the bladder cavity connected region includes: If the abnormal protrusion index corresponding to the edge pixel is greater than the preset abnormal threshold, then determine the edge pixel as an abnormal edge point; Form an abnormal protrusion edge with consecutive abnormal edge points.

7. The ultrasound imaging omics analysis method for the pathological characteristics of bladder tumors according to claim 1, characterized in that, The determination of the bladder tumor characteristic index corresponding to each abnormal protrusion edge according to the shape regularity of each abnormal protrusion edge and the gray-scale distribution between each abnormal protrusion edge and its corresponding normal edge line segment includes: Determine any one abnormal protrusion edge as a marked protrusion edge, and obtain the edge chain code sequence corresponding to the marked protrusion edge, where the edge chain code sequence corresponding to the marked protrusion edge is used to characterize the shape regularity of the marked protrusion edge; Determine the normal line of the center point of the marked protrusion edge as the target normal line corresponding to the marked protrusion edge, and determine the intersection point between the target normal line and the normal edge line segment corresponding to the marked protrusion edge as the target intersection point corresponding to the marked protrusion edge; Connect the center point of the marked protrusion edge and the target intersection point to obtain the target line segment corresponding to the marked protrusion edge, where the starting point of the target line segment is the center point of the marked protrusion edge, and the ending point of the target line segment is the target intersection point; Determine the bladder tumor characteristic index corresponding to the marked protrusion edge according to the edge chain code sequence corresponding to the marked protrusion edge and the gray-scale distribution on the target line segment corresponding to the marked protrusion edge.

8. The ultrasonic imaging omics analysis method for the pathological characteristics of bladder tumors according to claim 7, characterized in that The formula corresponding to the bladder tumor characteristic index corresponding to the marked protrusion edge is: ； Among them, is the bladder tumor characteristic index corresponding to the marked raised edge; is the normalization function; is the mean value of all chain codes in the edge chain code sequence corresponding to the marked raised edge; is the number of pixel points on the target line segment corresponding to the marked raised edge; is the serial number of the pixel points on the target line segment corresponding to the marked raised edge; is the th pixel point corresponding gray value on the target line segment corresponding to the marked raised edge; is the th pixel point corresponding gray value on the target line segment corresponding to the marked raised edge; is a preset factor greater than 0.

9. An ultrasound imaging omics analysis system for the pathological characteristics of bladder tumors, characterized in that, Including a processor and a memory, the processor is configured to process instructions stored in the memory to implement an ultrasonic imaging omics analysis method for a bladder tumor pathological feature according to any one of claims 1-8.

Images