Intelligent Processing Method for Ultrasonic Images in Obstetrics and Gynecology

Abstract

The present invention relates to the technical field of image filtering and enhancement, and particularly relates to an intelligent processing method for ultrasonic images in obstetrics and gynecology. The present invention acquires an abdominal wall ultrasonic image and performs region division to obtain a target region; further adjusts preset filtering window parameters according to the gray-scale fluctuation degree of pixel points in the target region; further divides the connected domains into tissue connected domains and noise connected domains according to the gray-scale uniformity of pixel points in each connected domain and the change of the gradient direction of edge pixel points; further corrects the initial window parameters according to the relative quantities of noise connected domains and tissue connected domains in the target region to obtain final window parameters; and finally filters the abdominal wall ultrasonic image according to the final window parameters of each image region. By analyzing the gray-scale fluctuation and noise performance of connected domains in the ultrasonic image region, the present invention adaptively determines the filtering window size, effectively removes speckle noise, and improves the recognition accuracy of ovarian cysts.

Description

Technical Field

[0001] The present invention relates to the technical field of image filtering and enhancement, and particularly relates to an intelligent processing method for ultrasonic images in obstetrics and gynecology. Background Art

[0002] Ultrasonic examination is an important examination method for clinical diseases in obstetrics and gynecology. With the continuous improvement of the structural technology of ultrasonic probes, the ultrasonic examination methods in obstetrics and gynecology have developed from simple transabdominal ultrasonic examination to various examination methods such as transvaginal ultrasonic examination and transuterine cavity ultrasonic examination.

[0003] Generally, there will be speckle noise in the unprocessed ultrasonic images in obstetrics and gynecology, which will seriously interfere with the normal diagnosis and analysis of ultrasonic images in obstetrics and gynecology. For example, polycystic ovary syndrome is a complex endocrine disorder disease, and its main diagnosis method is to count the number of small vesicles in the ovary in the ultrasonic image. Since the speckle noise is randomly distributed in the ultrasonic image, setting a smaller window may not effectively remove large-size speckle noise, while a larger window may cause over-smoothing and blurring, affecting the accuracy of the analysis results. Summary of the Invention

[0004] In order to solve the technical problem that the unreasonable setting of the filtering window affects the filtering effect of ultrasonic images in gynecology, the purpose of the present invention is to provide an intelligent processing method for ultrasonic images in obstetrics and gynecology, and the specific technical solution adopted is as follows:

[0005] Obtain the transabdominal ultrasonic image and perform regional division to obtain each image region of the transabdominal ultrasonic image; select each image region as the target region one by one;

[0006] Adjust the preset filtering window parameters according to the gray-scale fluctuation degree of the pixel points in the target region to obtain the initial window parameters of the target region; obtain the connected domains in the target region; obtain the noise possibility of each connected domain according to the gray-scale uniformity of the pixel points in each connected domain and the change of the gradient direction of the edge pixel points; divide the connected domains into tissue connected domains and noise connected domains according to the distribution of the noise possibility; correct the initial window parameters according to the relative quantity of the noise connected domains and the tissue connected domains in the target region to obtain the final window parameters of the target region;

[0007] Filter the transabdominal ultrasonic image according to the final window parameters of each image region.

[0008] Further, the method for obtaining the initial window parameters includes:

[0009] Obtain the gray-scale fluctuation intensity parameter of the target region according to the gray-scale range of the pixel points in the target region and the difference between the gray-scale value of each pixel point and the overall gray-scale;

[0010] Obtain the initial window parameter of the target region according to the gray-scale fluctuation intensity parameter and the preset filtering window parameter; both the gray-scale fluctuation intensity parameter and the preset filtering window parameter are positively correlated with the initial window parameter.

[0011] Further, the method for obtaining the noise probability includes:

[0012] Obtain the noise probability of each connected region according to the difference between the median and the mean of the gray-scale values of the pixel points in each connected region, and in combination with the change amplitude of the gradient direction of all adjacent edge pixel points; both the difference between the median and the mean of the gray-scale values and the change amplitude of the gradient direction are positively correlated with the noise probability.

[0013] Further, the method for dividing into tissue connected regions and noise connected regions includes:

[0014] Sort the noise probabilities of all the connected regions in the target region from small to large to form a sequence to be analyzed; divide the sequence to be analyzed into two parts according to the change of the noise probability in the sequence to be analyzed. The connected regions corresponding to the part where the maximum noise probability is located are noise connected regions, and the connected regions corresponding to the part where the minimum noise probability is located are tissue connected regions.

[0015] Further, the method for dividing the sequence to be analyzed into two parts according to the change of the noise probability in the sequence to be analyzed includes:

[0016] Sort the connected regions in the order of the noise probability in the sequence to be analyzed to obtain a connected region sequence; use the connected regions other than the two ends in the connected region sequence as the connected regions to be analyzed;

[0017] Obtain the boundary probability of each connected region to be analyzed according to the difference between the noise probabilities of the connected regions before and after each connected region to be analyzed; the difference between the noise probabilities of the connected regions before and after the connected region to be analyzed is positively correlated with the boundary probability; select the connected region to be analyzed with the largest boundary probability as the boundary point, and divide the boundary point and the connected regions before the boundary point into one part; divide the connected regions after the boundary point into another part.

[0018] Further, the method for obtaining the final window parameter includes:

[0019] In the target region, according to the difference between the number of noise connected components and the number of tissue connected components, the final window parameter of the target region is obtained by combining the initial window parameter; the difference between the number of noise connected components and the number of tissue connected components and the initial window parameter are both positively correlated with the final window parameter.

[0020] Further, connected components within the target region are extracted by using the Otsu threshold segmentation algorithm.

[0021] Further, the preset filtering window parameter is 5.

[0022] Further, the size of the image region is .

[0023] Further, median filtering is used to filter the abdominal wall ultrasound image.

[0024] The present invention has the following beneficial effects:

[0025] The present invention first obtains an abdominal wall ultrasound image and performs regional division to obtain a target region, so as to more carefully analyze the noise performance of different image regions, set more appropriate filtering parameters, and improve the filtering effect; further, the preset filtering window parameter is adjusted according to the gray level fluctuation degree of the pixel points within the target region to obtain the initial window parameter of the target region, initially adapting to the gray level fluctuation within the target region and preparing for further adjusting the filtering window size in the follow-up; further, the possibility that the connected component is speckle noise is evaluated from the perspectives of gray level uniformity and edge gradient direction change, providing a basis for classifying the connected components and finally correcting the filtering window in the follow-up; further, the connected components are divided into tissue connected components and noise connected components according to the distribution of noise possibility, facilitating further determining the noise situation within the target region and correcting the initial window parameter; further, according to the relative quantity of noise connected components and tissue connected components within the target region, the initial window parameter is corrected to obtain the final window parameter of the target region, improving the adaptability of the final window parameter, protecting the image structure edge, and enhancing the reliability of the analysis of obstetrics and gynecology images such as polycystic ovaries; finally, the abdominal wall ultrasound image is filtered according to the final window parameter of each image region. By analyzing the gray level fluctuation and noise performance of connected components within the ultrasound image region, the present invention adaptively determines the filtering window size, effectively removes speckle noise, protects the image detail texture, and improves the accuracy of ovarian cyst identification. Description of the Drawings

[0026] 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.

[0027] Figure 1 Flowchart of an ultrasonic image intelligent processing method for obstetrics and gynecology provided by an embodiment of the present invention;

[0028] Figure 2 Flowchart for obtaining an organizational connected domain and a noise connected domain provided by an embodiment of the present invention. Detailed implementation manners

[0029] To further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following combines the drawings and preferred embodiments to detail the specific implementation manners, structures, features, and effects of an ultrasonic image intelligent processing method for obstetrics and gynecology 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.

[0030] 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.

[0031] The following specifically describes the specific solution of an ultrasonic image intelligent processing method for obstetrics and gynecology provided by the present invention with reference to the drawings.

[0032] Please refer to Figure 1 , which shows the flowchart of an ultrasonic image intelligent processing method for obstetrics and gynecology provided by an embodiment of the present invention, specifically including:

[0033] Step S1: Obtain an abdominal wall ultrasonic image and perform region division to obtain each image region of the abdominal wall ultrasonic image; select each image region as the target region one by one.

[0034] In the embodiment of the present invention, first use an ultrasonic medical device to obtain the abdominal wall ultrasonic image of an obstetrics and gynecology patient. If the abdominal wall ultrasonic image is not a grayscale image, grayscale preprocessing is also required.

[0035] When using ultrasonic medical equipment to collect images of the abdominal wall, when sound waves interact with tissues or objects and return to the sensor, the interference and scattering effects of these sound waves will cause bright and dark spots to appear on the image, presenting random particles or spots, which is the speckle noise of the ultrasound image. Since the noise content of different image areas of the collected abdominal wall is different, the abdominal wall ultrasound image is divided into regions to obtain each image area of the abdominal wall ultrasound image, so as to analyze the noise performance of different image areas more carefully, set more appropriate filtering parameters, and improve the filtering effect.

[0036] As an example, the abdominal wall ultrasound image is evenly divided into image regions, and the size of each image region is .

[0037] It should be noted that the analysis method for each image region is consistent, and image regions are selected one by one as target regions. Only one target region is used as an example for explanation here, and no repetition is given. The implementer may also set other sizes for region division.

[0038] Step S2: Adjust the preset filter window parameters according to the grayscale fluctuation degree of the pixels in the target area to obtain the initial window parameters of the target area; obtain the connected domain in the target area; obtain the noise possibility of each connected domain according to the grayscale uniformity of the pixels in each connected domain and the gradient direction change of the edge pixels; divide the connected domain into tissue connected domain and noise connected domain according to the distribution of noise possibility; correct the initial window parameters according to the relative number of noise connected domain and tissue connected domain in the target area to obtain the final window parameters of the target area.

[0039] Considering that the grayscale fluctuation degree of pixels in the target area reflects the richness of the organizational content or the quantity characteristics of noise points in the target area, the preset filter window parameters are adjusted according to the grayscale fluctuation degree of pixels in the target area, and the initial window parameters of the target area are obtained to preliminarily adapt to the grayscale fluctuation in the target area, so as to prepare for the subsequent further adjustment of the filter window size.

[0040] Preferably, in one embodiment of the present invention, considering that speckle noise appears as randomly distributed bright and dark particles in an image, such as dark particles on a bright background (noise reduces local grayscale) in a high-echo area (such as ovarian stroma), and bright particles on a dark background (noise increases local grayscale) in a low-echo area (such as vesicular fluid), resulting in drastic grayscale fluctuations in the target area, requiring a larger filter window for smoothing and denoising; and the grayscale range of pixels in the area represents the grayscale fluctuation range, the larger the grayscale range, the more drastic the grayscale fluctuation; the greater the difference between the grayscale value of each pixel and the overall grayscale, the greater the grayscale discreteness and the more drastic the grayscale fluctuation;

[0041] Based on this, according to the gray range of pixel points in the target area and the difference between the gray value of each pixel point and the overall gray value, obtain the severe gray fluctuation parameter of the target area;

[0042] Obtain the initial window parameter of the target area according to the severe gray fluctuation parameter and the preset filtering window parameter; both the severe gray fluctuation parameter and the preset filtering window parameter are positively correlated with the initial window parameter.

[0043] As an example, take the product of the sum of the absolute values of the differences between the gray value of each pixel point in the target area and the mean of the gray values of all pixel points and the range of gray values as the gray fluctuation coefficient of the target area; take the ratio of the gray fluctuation coefficient of the target area to the sum of the gray fluctuation coefficients of all image areas as the severe gray fluctuation parameter of the target area, representing the degree of gray fluctuation of pixel points in the target area;

[0044] The preset filtering window parameter is 5, take the sum of the severe gray fluctuation parameter and the constant 1 as the initial adjustment coefficient, and take the product of the initial adjustment coefficient and the preset filtering window parameter as the initial window parameter of the target area.

[0045] As another example, considering that the variance of the gray values of pixel points in the target area can also reflect the degree of gray fluctuation, the greater the variance, the more severe the gray fluctuation. Multiply the sum of the absolute values of the differences between the gray value of each pixel point in the target area and the mean of the gray values of all pixel points, the variance of the gray values, and the range of gray values, and the product is used as the gray fluctuation coefficient of the target area.

[0046] In other embodiments of the present invention, the sum of the absolute values of the differences between the gray value of each pixel point in the target area and the mean of the gray values of all pixel points, the variance of the gray values, and the range of gray values can also be fused by addition or weighted summation; the gray fluctuation coefficient can also be normalized by linear normalization to obtain the severe gray fluctuation parameter.

[0047] Considering that when the richness of tissue content in the target area is relatively high, such as multiple adjacent small vesicles, blood vessels, tissue interfaces, etc., it will also cause pixel point gray fluctuations. To avoid confusion between rich tissue texture and dense speckle noise, resulting in unreasonable setting of the initial window parameter, further adjustment of the initial window parameter is also required.

[0048] Considering that the small vesicles in the ovary appear as multiple circular dark areas with clear edges and low echoes in the image, with uniform internal gray levels, while the speckle noise shows characteristics of uneven gray levels, blurred boundaries, and random gradient directions, and there are significant differences in the connected components formed by the two, the connected components within the target region are obtained; according to the gray level uniformity of the pixel points within each connected component and the change in the gradient direction of the edge pixel points, the noise possibility of each connected component is obtained, and the possibility that the connected component is speckle noise is evaluated from the perspectives of gray level uniformity and the change in the edge gradient direction, providing a basis for subsequent classification of the connected components and ultimately correcting the filtering window;

[0049] Considering that there are significant differences in the noise possibilities between the connected components of normal tissues such as small vesicles and the connected components of speckle noise, the connected components are divided into tissue connected components and noise connected components according to the distribution of the noise possibilities, preparing for subsequent filtering window correction.

[0050] In an embodiment of the present invention, the connected components within the target region are extracted by the Otsu threshold segmentation algorithm.

[0051] In other embodiments of the present invention, the implementer can also use methods such as the Sobel operator and morphological opening operation to extract the connected components, and the methods for obtaining the connected components are all well-known technical means to those skilled in the art and will not be elaborated here.

[0052] Preferably, in an embodiment of the present invention, considering that the speckle noise appears as isolated extremely high or extremely low gray level pixel points, a small number of extreme values will significantly affect the gray level mean but the median remains almost unchanged, and the difference between the median and the mean is large; while the gray levels within the connected components of normal tissues are uniform, and the median is close to the mean; at the same time, the gradient directions at the edges of the speckle noise are random, without a directional trend, and the change amplitude of the gradient direction is larger compared to normal tissues;

[0053] Based on this, according to the difference between the median and the mean of the gray level values of the pixel points within each connected component, combined with the change amplitude of the gradient directions of all adjacent edge pixel points, the noise possibility of each connected component is obtained; both the difference between the median and the mean of the gray level values and the change amplitude of the gradient direction are positively correlated with the noise possibility.

[0054] As an example, starting from the edge pixel point of the connected component closest to the lower left corner of the target region, in the clockwise direction, the absolute value of the difference in the gradient directions of any two adjacent edge pixel points is calculated, and the absolute value of the difference between the median and the mean of the gray level values of the pixel points within the connected component is multiplied by the average of the absolute values of the differences in the gradient directions of all adjacent edge pixel points, and the product is linearly normalized to be used as the noise possibility of the corresponding connected component.

[0055] Among them, the difference between the median and the mean is reflected by the absolute value of the difference, showing the gray-scale uniformity of the pixel points within the connected domain. The smaller the absolute value of the difference between the median and the mean, the stronger the gray-scale uniformity. Similarly, the change amplitude of the gradient directions of adjacent pixel points is shown by the absolute value of the difference, and then the change of the gradient directions of all adjacent edge pixel points is reflected by averaging, and the noise possibility of the connected domain is obtained by multiplying.

[0056] In other embodiments of the present invention, the implementer can also fuse the absolute value of the difference between the median and the mean of the gray-scale values of the pixel points within the connected domain and the average value of the absolute values of the differences of the gradient directions of all adjacent edge pixel points by addition or weighted summation; the average absolute deviation, variance, etc. of the gray-scale values can also be added to measure the gray-scale discrete characteristics and analyze the gray-scale uniformity, which will not be elaborated here.

[0057] Preferably, in an embodiment of the present invention, please refer to Figure 2 , which shows a flowchart for obtaining an organization connected domain and a noise connected domain provided by an embodiment of the present invention, specifically including:

[0058] Step S201: Sort the noise possibilities of all connected domains within the target area from small to large to form a sequence to be analyzed.

[0059] Considering that it is easier to analyze the change of the noise possibility after sorting the noise possibilities, a sequence to be analyzed is first constructed, and the distribution of the noise possibility is analyzed with the help of the sequence to be analyzed.

[0060] Step S202: According to the change of the noise possibility in the sequence to be analyzed, divide the sequence to be analyzed into two parts. The connected domain corresponding to the part where the maximum noise possibility is located is the noise connected domain, and the connected domain corresponding to the part where the minimum noise possibility is located is the organization connected domain.

[0061] Considering that the noise possibility of the connected domain of normal tissue is small and the noise possibility of the connected domain of speckle noise is large, showing an obvious boundary feature in the sequence to be analyzed. Therefore, according to the change of the noise possibility in the sequence to be analyzed, the sequence to be analyzed is divided into two parts. At the same time, in the two segmented sequences, the maximum noise possibility is most likely to correspond to the speckle noise. Therefore, the connected domain corresponding to the part where the maximum noise possibility is located is the noise connected domain, and the connected domain corresponding to the part where the minimum noise possibility is located is the organization connected domain.

[0062] In an embodiment of the present invention, to analyze the boundary feature of the sequence to be analyzed, the sequence to be analyzed is divided into two parts, the connected domains are sorted according to the sorting order of the noise possibilities in the sequence to be analyzed to obtain a connected domain sequence; the connected domains in the connected domain sequence except for the two ends are used as the connected domains to be analyzed;

[0063] Considering that when the difference in noise possibility between the connected regions before and after the connected region to be analyzed is the largest, it indicates that the boundary feature is the most obvious. Therefore, according to the difference in noise possibility between the connected regions before and after each connected region to be analyzed, the boundary possibility of each connected region to be analyzed is obtained; the difference in noise possibility between the connected regions before and after the connected region to be analyzed is positively correlated with the boundary possibility.

[0064] As an example, the connected region itself to be analyzed and the connected regions before it are regarded as the connected regions before the connected region to be analyzed, and the connected regions after the connected region to be analyzed are regarded as the connected regions before the connected region to be analyzed;

[0065] The absolute value of the difference between the mean noise possibility of the connected regions before the connected region to be analyzed and the mean noise possibility of the connected regions after the connected region to be analyzed is used as the first difference parameter; the absolute value of the difference between the noise possibility of the connected region to be analyzed and the noise possibility of the adjacent next connected region is used as the second difference parameter; the product of the first difference parameter and the second difference parameter is used as the boundary possibility of the connected region to be analyzed.

[0066] Among them, in the form of the mean value, the overall characteristics of the noise possibility of the connected regions before and after the connected region to be analyzed are respectively represented by the mean noise possibility of the connected regions. In the way of the absolute value of the difference, the first difference parameter shows the difference in noise possibility before and after the connected region to be analyzed from an overall perspective; then the second difference parameter shows the sudden increase amplitude of the noise possibility before and after the connected region to be analyzed, showing the difference in noise possibility before and after the connected region to be analyzed from a local perspective; finally, they are fused by multiplication to obtain the boundary possibility.

[0067] Select the connected region to be analyzed with the largest boundary possibility as the boundary point, and divide the boundary point and the connected regions before the boundary point into one part (organized connected regions); divide the connected regions after the boundary point into another part (noise connected regions).

[0068] In another embodiment of the present invention, considering that the maximum value in the first-order difference sequence of the sequence to be analyzed corresponds to the most obvious sudden increase change in noise possibility, which thus corresponds to the boundary between the noise possibilities of the two types of connected regions, the sequence to be analyzed is segmented from the position of the maximum value of the elements in the first-order difference sequence.

[0069] In other embodiments of the present invention, the implementer can also adopt the method of threshold screening, set the noise possibility threshold, regard the connected regions smaller than the noise possibility threshold as tissue connected regions, and regard the connected regions greater than or equal to the noise possibility threshold as noise connected regions; among them, several (such as 100) ultrasonic images can be selected in advance, the tissue connected regions and noise connected regions are manually marked, the noise possibility is calculated, and the position with the least intersection of the two distributions is selected as the threshold, for example, the noise possibility of 95% of the tissue connected regions is less than the noise possibility threshold, and the noise possibility of 80% of the noise connected regions is greater than or equal to the noise possibility threshold.

[0070] The K-means clustering algorithm can also be used to set K = 2 to classify the noise possibility, so as to classify the connected regions.

[0071] Furthermore, considering that the relative quantity of noise connected regions and tissue connected regions in the target area reflects the dominant position of the two types of connected regions, it is judged whether the target area is dominated by normal tissue or noise, so as to dynamically adjust the initial window parameters. Therefore, according to the relative quantity of noise connected regions and tissue connected regions in the target area, the initial window parameters are corrected to obtain the final window parameters of the target area, improving the fitness of the final window parameters, protecting the image structure edges, and enhancing the reliability of the analysis of obstetrics and gynecology images such as polycystic ovaries.

[0072] Preferably, in one embodiment of the present invention, considering that the more the number of noise connected regions is compared with the number of tissue connected regions, the stronger the dominant position of the noise is, the filtering window should be enlarged to enhance smoothing and suppress the random speckle noise as much as possible; on the contrary, the fewer the number of noise connected regions is compared with the number of tissue connected regions, the more organizational structures there are in the area, and it cannot be over-smoothed. The filtering window should be reduced to avoid damaging important structures such as edges and blurring vesicles.

[0073] Based on this, in the target area, the final window parameters of the target area are obtained according to the difference between the number of noise connected regions and the number of tissue connected regions, in combination with the initial window parameters; the difference between the number of noise connected regions and the number of tissue connected regions and the initial window parameters are both positively correlated with the final window parameters.

[0074] As an example, the difference between the number of noise connected regions and the number of tissue connected regions is used as the numerator, the sum of the number of noise connected regions and the number of tissue connected regions is used as the denominator, and the sum of the fractional ratio and the constant 1 is used as the final correction coefficient; after rounding up the product of the final correction coefficient and the initial window parameters to the nearest odd integer, the rounded result is used as the final window parameters.

[0075] It should be noted that the minimum value of the final window parameters is 3. When it is less than 3, it is set to 3, and the corresponding filtering window size is .

[0076] In another embodiment of the present invention, the initial window parameters are further corrected by integrating the area angle. As an example, the difference between the number of noise connected regions and the number of tissue connected regions is used as the numerator, the sum of the number of noise connected regions and the number of tissue connected regions is used as the denominator, and the fractional ratio is used as the first correction factor;

[0077] The difference between the total area of the noise connected regions and the total area of the tissue connected regions is used as the numerator, the total area of the noise connected regions and the total area of the tissue connected regions is used as the denominator, and the fractional ratio is used as the second correction factor;

[0078] The first correction factor and the second correction factor are weighted and summed with weights of 0.8 and 0.2, and the sum of the weighted sum result and the constant 1 is used as the final correction coefficient; among them, since the area of normal tissue is usually larger than that of the noise connected regions, the noise dominance will not be too prominent from the area angle, and the number angle is more sensitive, so the weight given to the second correction factor is smaller.

[0079] Step S3: Filter the abdominal wall ultrasound image according to the final window parameters of each image region.

[0080] By analyzing the gray scale fluctuation and noise evaluation of the abdominal wall ultrasound image in regions, the filtering window is dynamically corrected to obtain the most suitable final window parameters for each image region, balancing the requirements of protecting image details and filtering out noise interference. Finally, the abdominal wall ultrasound image is filtered according to the final window parameters of each image region.

[0081] Preferably, in an embodiment of the present invention, considering that the speckle noise in the ultrasound image is random dot noise, median filtering is naturally suitable for eliminating sharp isolated points while retaining edge information, which exactly matches the characteristics of the ultrasound image. Therefore, the abdominal wall ultrasound image is filtered by median filtering to provide reliable image data support for relevant personnel.

[0082] It should be noted that in other embodiments of the present invention, the implementer can also adopt other window filtering methods such as weighted median filtering. Both it and median filtering are well-known technical means to those skilled in the art, and will not be elaborated here.

[0083] In summary, in view of the technical problem that the unreasonable setting of the filtering window affects the filtering effect of gynecological ultrasound images, the present invention proposes an intelligent processing method for ultrasonic images in obstetrics and gynecology. The present invention acquires an abdominal wall ultrasound image and performs region division to obtain a target region; further adjusts preset filtering window parameters according to the gray-scale fluctuation degree of pixel points in the target region; further divides the connected domain into a tissue connected domain and a noise connected domain according to the gray-scale uniformity of pixel points in each connected domain and the change in the gradient direction of edge pixel points; further corrects the initial window parameters according to the relative quantities of the noise connected domain and the tissue connected domain in the target region to obtain final window parameters; and finally filters the abdominal wall ultrasound image according to the final window parameters of each image region. By analyzing the gray-scale fluctuation and noise performance of connected domains in the ultrasonic image region, the present invention adaptively determines the filtering window size, effectively removes speckle noise, and improves the recognition accuracy of ovarian vesicles.

[0084] It should be noted that the above sequence of embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0085] Each embodiment in this specification is described in a progressive manner, and the same or similar parts among the embodiments can be referred to each other. The key point of each embodiment is to illustrate the differences from other embodiments.

Claims

1. An intelligent processing method for ultrasonic images in obstetrics and gynecology, characterized in that, The method includes: Obtaining an abdominal wall ultrasound image and performing regional division to obtain each image region of the abdominal wall ultrasound image; successively selecting each of the image regions as a target region; Adjusting preset filtering window parameters according to the gray-scale fluctuation degree of pixel points in the target region to obtain initial window parameters of the target region; obtaining connected regions in the target region; obtaining the noise possibility of each connected region according to the gray-scale uniformity of pixel points in each connected region and the change in the gradient direction of edge pixel points; dividing the connected regions into tissue connected regions and noise connected regions according to the distribution of the noise possibilities; correcting the initial window parameters according to the relative quantities of the noise connected regions and the tissue connected regions in the target region to obtain final window parameters of the target region; Filtering the abdominal wall ultrasound image according to the final window parameters of each image region; The method for dividing into tissue connected regions and noise connected regions includes: Sorting the noise possibilities of all the connected regions in the target region from small to large to form a sequence to be analyzed; dividing the sequence to be analyzed into two parts according to the change in the noise possibilities in the sequence to be analyzed, where the connected regions corresponding to the part where the maximum noise possibility is located are noise connected regions, and the connected regions corresponding to the part where the minimum noise possibility is located are tissue connected regions; The method for dividing the sequence to be analyzed into two parts according to the change in the noise possibilities in the sequence to be analyzed includes: Sorting the connected regions in the order of the noise possibilities in the sequence to be analyzed to obtain a connected region sequence; using the connected regions other than the two ends in the connected region sequence as connected regions to be analyzed; Obtaining the demarcation possibility of each connected region to be analyzed according to the difference in the noise possibilities of the connected regions before and after each connected region to be analyzed; the difference in the noise possibilities of the connected regions before and after the connected region to be analyzed is positively correlated with the demarcation possibility; selecting the connected region to be analyzed with the largest demarcation possibility as the demarcation point, and dividing the demarcation point and the connected regions before the demarcation point into one part; dividing the connected regions after the demarcation point into another part.

2. The intelligent processing method for ultrasonic images used in obstetrics and gynecology according to claim 1, wherein The method for obtaining the initial window parameters includes: Obtaining a gray-scale fluctuation intensity parameter of the target region according to the gray-scale range of pixel points in the target region and the difference between the gray-scale value of each pixel point and the overall gray scale; Obtaining the initial window parameters of the target region according to the gray-scale fluctuation intensity parameter and the preset filtering window parameters; both the gray-scale fluctuation intensity parameter and the preset filtering window parameters are positively correlated with the initial window parameters.

3. The intelligent processing method for ultrasonic images used in obstetrics and gynecology according to claim 1, characterized in that, The method for obtaining the noise possibility includes: Obtaining the noise possibility of each connected region according to the difference between the median value and the mean value of the gray-scale values of pixel points in each connected region, in combination with the change amplitude of the gradient directions of all adjacent edge pixel points; the difference between the median value and the mean value of the gray-scale values, and the change amplitude of the gradient directions are both positively correlated with the noise possibility.

4. The ultrasonic image intelligent processing method for obstetrics and gynecology according to claim 1, characterized in that The method for obtaining the final window parameters includes: In the target region, according to the difference between the number of noise connected components and the number of tissue connected components, the final window parameters of the target region are obtained by combining the initial window parameters; the difference between the number of noise connected components and the number of tissue connected components and the initial window parameters are both positively correlated with the final window parameters.

5. A method for intelligent processing of ultrasonic images for obstetrics and gynecology according to claim 1, characterized in that, Connected components within the target region are extracted by the Otsu threshold segmentation algorithm.

6. The ultrasonic image intelligent processing method for obstetrics and gynecology according to claim 1, characterized in that, The preset filtering window parameter is 5.

7. A method for intelligent processing of ultrasonic images for obstetrics and gynecology according to claim 1, characterized in that, The size of the image area is .

8. A method for intelligent processing of ultrasonic images for obstetrics and gynecology according to claim 1, characterized in that Median filtering is used to filter the abdominal wall ultrasound image.

Images