An intelligent processing method for cardiac ultrasound images
By dividing areas, calculating cardiac expression and screening abnormal pixel points in cardiac ultrasound images, the image blur problem caused by inappropriate probe placement is solved, and the image quality of cardiac ultrasound images is improved.
Patent Information
- Application Number
- CN202411338232.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-25
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-09-25
AI Technical Summary
During the acquisition of cardiac ultrasound images, the probe position is inappropriate due to factors such as the patient's body shape and respiratory movement, resulting in blurred part of the cardiac ultrasound images, which reduces the image quality.
By obtaining a binary image of the heart ultrasound, dividing the regions and calculating the heart performance, filtering out the heart region and abnormal pixel points, and correcting the image according to the correction coefficient of the abnormal pixel points to improve the image quality.
It improves the accuracy of cardiac area recognition, avoids interference from non-cardiac areas, and enhances the image quality of cardiac ultrasound images.
Smart Images

Figure CN118864460B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of image processing, and in particular to an intelligent processing method for cardiac ultrasonic images. Background Art
[0002] Cardiac ultrasound imaging is an indispensable and important examination method in the diagnosis and treatment of cardiovascular diseases. It has the advantages of real-time imaging, economy, speed, non-invasiveness, and no radiation. By utilizing the propagation and reflection characteristics of ultrasound in human tissues, the image information of the internal structure of the heart is obtained by receiving and processing the reflected ultrasound signals. This technology provides doctors with a unique window to gain an in-depth understanding of the structure and function of the heart, which is of great significance for the diagnosis and treatment of heart diseases. The heart is a complex three-dimensional structure, and different sections require different probe positions and angles to be displayed clearly. In the process of using cardiac ultrasound imaging equipment to collect cardiac ultrasound images of patients, factors such as the patient's body shape (such as obesity, chest wall thickness), respiratory movement, and heart position may affect the placement of the probe, making the probe placement inappropriate, making the acquired cardiac ultrasound image partially blurred, and reducing the image quality of the cardiac ultrasound image. Summary of the invention
[0003] The invention provides an intelligent processing method for cardiac ultrasonic images to solve the existing problems.
[0004] The intelligent processing method of cardiac ultrasound images of the present invention adopts the following technical scheme:
[0005] An embodiment of the present invention provides an intelligent processing method for cardiac ultrasound images, the method comprising the following steps:
[0006] Acquire a cardiac ultrasound image and a cardiac ultrasound binary image of the cardiac ultrasound image;
[0007] According to the gray value of each pixel point on each column in the binary cardiac ultrasound image, the binary cardiac ultrasound image is divided into a plurality of regions; according to the gray value of each pixel point on each column in each region and the number of columns in each region, the cardiac expression degree of each region is obtained; according to the cardiac expression degree of each region, a plurality of cardiac regions are screened out from all regions;
[0008] The area except the heart area in the cardiac ultrasound binary image is equally divided into several wholes; according to the distance between each whole and each heart area, the gray value of each pixel in each whole and the gray value of each pixel in each heart area, several abnormal pixels are screened out from all the wholes; according to the gray value of each abnormal pixel and the gray value of each pixel in the heart area closest to each abnormal pixel, a correction coefficient of each abnormal pixel is obtained; according to the correction coefficient of each abnormal pixel, a processed cardiac ultrasound image is obtained.
[0009] Furthermore, the method of dividing the cardiac ultrasound binary image into a plurality of regions according to the grayscale value of each pixel point in each column of the cardiac ultrasound binary image comprises the following specific steps:
[0010] In the cardiac ultrasound binary image, the column with the largest number of pixels with a gray value of 1 in all columns is used as the region division column, and the cardiac ultrasound binary image is divided into several regions.
[0011] Furthermore, the cardiac expression degree of each region is obtained according to the gray value of each pixel on each column in each region and the number of columns in each region, and the specific steps include the following:
[0012] Each area from top to bottom The area in the row, from bottom to top The area in the row, from left to right The area in the column, from right to left The areas in the columns are recorded as target areas, where A first quantity threshold is preset;
[0013] The pixels with continuous adjacent gray values of 1 in each target area form a connected domain, and several connected domains are obtained;
[0014] When the number of pixels in the connected domain in each target area is greater than or equal to When the target area is recorded as the main target area, A second quantity threshold is preset;
[0015] According to the number of main target areas in each area, the gray value of each pixel on each column in each area, and the number of columns in each area, the cardiac expression degree of each area is obtained.
[0016] Furthermore, according to the number of main target areas in each area, the gray value of each pixel on each column in each area, and the number of columns in each area, a specific calculation formula for obtaining the cardiac expression degree of each area is:
[0017]
[0018] in, Indicates The cardiac performance of each region, Indicates The maximum value of the number of pixels between the first pixel with a grayscale value of 0 and the last pixel with a grayscale value of 0 in all columns in the region, Indicates Region The number of pixels with grayscale value 0 in the column, Indicates The number of columns in the region, represents the linear normalization function, For the The number of primary target areas in the area.
[0019] Furthermore, the method of selecting a plurality of cardiac regions from all regions according to the cardiac expression of each region includes the following specific steps:
[0020] The cardiac expression indexes of all regions are arranged in descending order to obtain a cardiac expression index sequence of all regions, and the regions corresponding to the first two values in the cardiac expression index sequence are recorded as cardiac regions.
[0021] Furthermore, the method of screening out a number of abnormal pixels from all the entities according to the distance between each entity and each heart region, the grayscale value of each pixel in each entity, and the grayscale value of each pixel in each heart region includes the following specific steps:
[0022] Jordi The whole is in the In the region, The area is the distance the nearest region to the whole;
[0023] Jordi The whole is not all in the In the area, calculate the The center point of the whole and the existence of The distance between the center points of each region of the whole will be The area with the shortest distance to the center point of the whole is recorded as the distance the nearest region to the whole;
[0024] The degree of abnormal performance of each whole is obtained according to the area closest to each whole, the gray value of each pixel in each whole, and the gray value of each pixel in each heart area;
[0025] The whole with an abnormal performance degree greater than or equal to a preset judgment threshold is recorded as an abnormal whole, and the pixel points in the abnormal whole are recorded as abnormal pixels.
[0026] Furthermore, the specific calculation formula for obtaining the abnormal performance degree of each entity according to the area closest to each entity, the grayscale value of each pixel in each entity, and the grayscale value of each pixel in each heart area is:
[0027]
[0028] in, Performance The overall degree of abnormal performance, Indicates the distance The cardiac expression of the nearest region of the whole, Indicates The distance between the center point of the whole and the center point of the nearest heart region, Indicates The number of pixels with gray value 0 in the whole Indicates The number of pixels in the whole, Represents the grayscale mean of all pixels in all heart regions; Indicates The grayscale mean of all pixels in the whole is the linear normalization function, represents the absolute value function.
[0029] Furthermore, the correction coefficient of each abnormal pixel is obtained according to the gray value of each abnormal pixel and the gray value of each pixel in the heart area closest to each abnormal pixel, and the specific steps include the following:
[0030] According to the one-to-one correspondence between the cardiac ultrasound image and the pixel points in the cardiac ultrasound binary image, the first The specific calculation formula for the correction coefficient of an abnormal pixel is:
[0031]
[0032] in, Indicates The correction coefficient of abnormal pixels, Indicates The overall abnormal performance degree of the abnormal pixel point, Indicates The gray value of the pixel corresponding to each abnormal pixel in the cardiac ultrasound image; Indicates the distance The center point of the heart area closest to the abnormal pixel point corresponds to the gray value of the pixel point in the cardiac ultrasound image; Indicates the distance The grayscale mean of all pixels in the heart area closest to the abnormal pixel corresponding to the pixel in the cardiac ultrasound image, Indicates The abnormal pixel point and the distance The distance between the center points of the heart area closest to the abnormal pixel points, is a linear normalization function.
[0033] Furthermore, the processing of the cardiac ultrasound image according to the correction coefficient of each abnormal pixel point includes the following specific steps:
[0034] Determine abnormal pixel points in the cardiac ultrasound image and correction coefficients of the abnormal pixel points according to a one-to-one correspondence between the cardiac ultrasound image and the pixel points in the cardiac ultrasound binary image;
[0035] Obtaining a corrected grayscale value of each abnormal pixel in the cardiac ultrasound image according to a correction coefficient of each abnormal pixel in the cardiac ultrasound image and a grayscale value of each abnormal pixel;
[0036] The gray value of each abnormal pixel in the cardiac ultrasound image is replaced with the corrected gray value to obtain a processed cardiac ultrasound image.
[0037] Furthermore, the step of obtaining the corrected grayscale value of each abnormal pixel in the cardiac ultrasound image according to the correction coefficient of each abnormal pixel in the cardiac ultrasound image and the grayscale value of each abnormal pixel includes the following specific steps:
[0038] Calculate the first The correction coefficient of the abnormal pixel point is The product of the grayscale values of the abnormal pixel points is added to the grayscale value of the abnormal pixel point in the cardiac ultrasound image. The sum of the grayscale values of the abnormal pixels is rounded upwards and recorded as the value of the first abnormal pixel in the cardiac ultrasound image. The gray value of the abnormal pixel after correction.
[0039] The beneficial effects of the technical solution of the present invention are:
[0040] In the embodiment of the present invention, according to the gray value of each pixel on each column in each region of the cardiac ultrasound binary image and the number of columns in each region, several cardiac regions are screened out from all regions; the region is divided according to the gray value of each pixel on each column, which improves the accuracy of cardiac region recognition and avoids the interference of non-cardiac regions in the cardiac ultrasound image processing process. The region other than the cardiac region in the cardiac ultrasound binary image is equally divided into several wholes, and according to the distance between each whole and each cardiac region, the gray value of each pixel in each whole and the gray value of each pixel in each cardiac region, several abnormal pixels are screened out from all pixels; according to the gray value of each abnormal pixel and the gray value of each pixel in the cardiac region closest to each abnormal pixel, the processed cardiac ultrasound image is obtained. So far, the present invention identifies the position of the heart through the characteristic performance of the heart. Further analysis is performed to obtain the distribution performance of abnormal pixels on the image, and correction is performed to improve the image quality of the cardiac ultrasound image. BRIEF DESCRIPTION OF THE DRAWINGS
[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0042] Figure 1 A flowchart of the steps of an intelligent processing method for cardiac ultrasound images of the present invention;
[0043] Figure 2 It is a schematic diagram of the cardiac ultrasound binary image after dividing the regions in this embodiment. DETAILED DESCRIPTION
[0044] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the intelligent processing method of cardiac ultrasound images proposed by the present invention, its specific implementation, structure, features and effects, in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.
[0045] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
[0046] The specific scheme of the intelligent processing method of cardiac ultrasound images provided by the present invention is described in detail below with reference to the accompanying drawings.
[0047] See also Figure 1 , which shows a flowchart of a method for intelligent processing of cardiac ultrasound images provided by an embodiment of the present invention, the method comprising the following steps:
[0048] Step S001: Acquire a cardiac ultrasound image and a cardiac ultrasound binary image of the cardiac ultrasound image.
[0049] The ultrasonic image acquisition device is used to acquire the patient's cardiac ultrasound image. The cardiac ultrasound image is cone-shaped. Figure 2 The area outside the middle cone-shaped cardiac ultrasound image is a black background area, and the analysis in this embodiment does not include the black background area.
[0050] It should be noted that: median filtering is used to denoise the acquired cardiac ultrasound images of the patient to reduce the influence of abnormal factors during the acquisition process.
[0051] When using ultrasonic imaging equipment to collect the patient's cardiac ultrasound image, the imaging probe position may not be in the optimal position due to the patient's body shape, breathing conditions, etc., resulting in insufficient penetration and some blurred areas on the cardiac ultrasound image. It is impossible to determine whether they are heart areas, which affects the doctor's diagnosis of the patient's heart. These blurred areas need to be further analyzed to analyze whether they should be considered as heart areas.
[0052] The heart region on the cardiac ultrasound image has a relatively obvious shape.
[0053] The heart region in the cardiac ultrasound image is segmented from other regions using threshold segmentation to obtain a cardiac ultrasound binary image of the cardiac ultrasound image.
[0054] Among them, the threshold segmentation uses the Otsu algorithm, which is a well-known technology. The grayscale value of the pixel points in the heart area segmented by the threshold is smaller than that in other areas. The grayscale value of the pixel points in the heart area is set to 0, and the grayscale value of the pixel points in other areas is set to 1.
[0055] The overall appearance of the cardiac ultrasound binary image is generally cone-shaped, which represents the boundary of the cardiac ultrasound segmentation image. Due to the shape of the heart, there will always be pixel points with a gray value of 1 at the boundary of the cardiac ultrasound binary image.
[0056] In the binary cardiac ultrasound image, the roughly enclosed pixels with a grayscale value of 0 are the pixels of the heart area. There are many pixels with grayscale values of 0 or 1 scattered near some pixels connected to the heart area. These pixels may themselves belong to the pixels in the heart area, but due to incorrect probe position and insufficient penetration during the acquisition process, some pixels on the image are blurred and need to be corrected.
[0057] Step S002: Divide the cardiac ultrasound binary image into several regions according to the grayscale value of each pixel on each column in the cardiac ultrasound binary image; obtain the cardiac expression degree of each region according to the grayscale value of each pixel on each column in each region and the number of columns in each region; and select several cardiac regions from all regions according to the cardiac expression degree of each region.
[0058] The heart region normally appears as two raised ovals, and it is determined whether a pixel point with a gray value of 0 on the cardiac ultrasound binary image is a pixel point on the heart region.
[0059] Since the pixel distribution in the heart area is convex, it is surrounded by pixels with a grayscale value of 1 on both sides. The column where there are more consecutive pixels with a grayscale value of 1 is used as the area division column (normally it will be divided into 2 areas). If it is divided into two areas, both are heart areas.
[0060] In the cardiac ultrasound binary image, the column with the largest number of pixels with a gray value of 1 in all columns is used as the region division column, and the cardiac ultrasound binary image is divided into several regions.
[0061] It should be noted that: when there is only one column with the largest number of pixels with grayscale values of 1 in all columns, the cardiac ultrasound binary image will be divided into two regions, and both regions are cardiac regions. When there are multiple columns with the largest number of pixels with grayscale values of 1 in all columns, the cardiac ultrasound binary image will be divided into multiple regions. Figure 2 As shown, Figure 2 Schematic diagram of the cardiac ultrasound binary image after it is divided into regions in this embodiment, wherein the cardiac ultrasound binary image is divided into three regions, namely region 1, region 2 and region 3.
[0062] If it is divided into multiple areas, it is necessary to further determine which two areas are the heart areas.
[0063] Each area from top to bottom The area in the row, from bottom to top The area in the row, from left to right The area in the column, from right to left The areas in the columns are recorded as target areas.
[0064] The pixels with continuous adjacent gray value of 1 in each target area form a connected domain, and several connected domains are obtained.
[0065] When the number of pixels in the connected domain in each target area is greater than or equal to When the target area is 0, the target area is recorded as the main target area.
[0066] It should be noted that: in this embodiment, the first quantity threshold is preset is 10, and the second quantity threshold is preset The value is 20, and this is used as an example for description.
[0067] First For example, the first The calculation formula of the cardiac expressivity of each region is:
[0068]
[0069] in, Indicates The cardiac performance of each region, Indicates The maximum value of the number of pixels between the first pixel with a grayscale value of 0 and the last pixel with a grayscale value of 0 in all columns in the region. Indicates Region The number of pixels with grayscale value 0 in the column, Indicates The number of columns in the region, represents the linear normalization function, For the The number of primary target areas.
[0070] It should be noted that the heart area is elliptical in shape, its height (i.e., the difference between the upper and lower endpoints) should be large, and there are continuous pixels with grayscale values of 0 in the vertical direction. To calculate the The heart expression of the region should be wrapped around the heart region, so the first The number of primary target areas Calculate restrictions.
[0071] The cardiac expression indexes of all regions are arranged in descending order to obtain a cardiac expression index sequence of all regions, and the regions corresponding to the first two values in the cardiac expression index sequence are recorded as cardiac regions.
[0072] Step S003: Divide the area except the heart area in the cardiac ultrasound binary image into several wholes; select several abnormal pixels from all the wholes according to the distance between each whole and each heart area, the grayscale value of each pixel in each whole and the grayscale value of each pixel in each heart area; obtain a correction coefficient for each abnormal pixel according to the grayscale value of each abnormal pixel and the grayscale value of each pixel in the heart area closest to each abnormal pixel; obtain a processed cardiac ultrasound image according to the correction coefficient of each abnormal pixel.
[0073] However, since the probe position is usually not the optimal position, the penetration is insufficient, resulting in blurred pixels in some areas, which is manifested in the presence of cross-distributed pixels with gray values of 0 or 1 on the cardiac ultrasound binary image. These areas are generally distributed at the boundaries of the heart area, so further analysis of these pixels is required.
[0074] The area except the heart area in the cardiac ultrasound binary image is equally divided into several wholes.
[0075] It should be noted that: the size of each whole is 3*3, and this is taken as an example for description.
[0076] Jordi The whole is in the In the region, The area is the distance the nearest region to the whole;
[0077] Jordi The whole is not all in the In the area, calculate the The center point of the whole and the existence of The distance between the center points of each region of the whole will be The area with the shortest distance to the center point of the whole is recorded as the distance The nearest area to the whole.
[0078] First Take the whole as an example, then The calculation formula for the degree of abnormal performance of an entire group is:
[0079]
[0080] in, Performance The overall degree of abnormal performance, Indicates the distance The cardiac expressivity of the region closest to the whole. Indicates The distance between the center point of the whole and the center point of the nearest heart region, Indicates The number of pixels with gray value 0 in the whole Indicates The number of pixels in the whole, Represents the grayscale mean of all pixels in all heart regions; Indicates The grayscale mean of all pixels in the whole is the linear normalization function, represents the absolute value function.
[0081] It should be noted that: when the cardiac ultrasound image appears blurred due to insufficient penetration, there will be some areas on the cardiac ultrasound binary image where pixels with grayscale values of 0 and 1 are cross-distributed. The more pixels with a grayscale value of 0, the closer they are to the heart area, and the smaller the difference between their grayscale mean and the grayscale mean of the heart area, the greater the possibility that they are abnormal as a whole (that is, the pixels should be in the heart area).
[0082] A judgment threshold is preset, and the whole with an abnormal performance degree greater than or equal to the preset judgment threshold is recorded as an abnormal whole, and the pixel points in the abnormal whole are recorded as abnormal pixels.
[0083] It should be noted that: in this implementation, the preset judgment threshold is 0.7, and this is used as an example for description.
[0084] The distribution of abnormal pixels on the cardiac ultrasound binary image is obtained in the above process. Since these abnormal pixels should be divided into cardiac areas, but the pixels in these areas are blurred due to insufficient penetration of the equipment, and there are differences in grayscale values, it is necessary to correct the grayscale values of these abnormal pixels. Since the specific performance of each cardiac area is different, its grayscale value performance is also different. The abnormal pixels distributed around it should have similar grayscale values to the cardiac areas that are closer to it.
[0085] According to the one-to-one correspondence between the cardiac ultrasound image and the pixel points in the cardiac ultrasound binary image, the first The calculation formula for the correction coefficient of an abnormal pixel is:
[0086]
[0087] in, Indicates The correction coefficient of abnormal pixels, Indicates The overall abnormal performance degree of the abnormal pixel point, Indicates The gray value of the pixel corresponding to each abnormal pixel in the cardiac ultrasound image; Indicates the distance The center point of the heart area closest to the abnormal pixel point corresponds to the gray value of the pixel point in the cardiac ultrasound image; Indicates the distance The grayscale mean of all pixels in the heart area closest to the abnormal pixel corresponding to the pixel in the cardiac ultrasound image, Indicates The abnormal pixel point and the distance The distance between the center points of the heart area closest to the abnormal pixels. is a linear normalization function.
[0088] It should be noted that: for abnormal pixels, the greater the degree of abnormal performance of the whole, the greater the degree of correction required for the corresponding abnormal pixel; and if the distance between the abnormal pixel and the center point of the heart area closest to it is smaller, the grayscale value difference (including the average grayscale value difference and the individual grayscale value difference) is greater, which means that the grayscale performance difference between it and the heart area is greater, and the degree of correction required is greater.
[0089] In the above process, we obtained The correction coefficient of the abnormal pixel point. According to the calculated correction coefficient of the abnormal pixel point, each abnormal pixel point is processed with a specific grayscale value. The size of the correction coefficient can reflect the degree of correction of each abnormal pixel point. The larger the correction coefficient, the greater the degree of correction required for this abnormal pixel point. According to the calculated correction coefficient, each abnormal pixel point is processed to obtain the grayscale value of each abnormal pixel point after processing.
[0090] According to the one-to-one correspondence between the cardiac ultrasonic image and the pixel points in the cardiac ultrasonic binary image, the abnormal pixel points in the cardiac ultrasonic image and the correction coefficients of the abnormal pixel points are determined.
[0091] In cardiac ultrasound images For example, the abnormal pixel point in the cardiac ultrasound image The calculation formula for the gray value of the corrected abnormal pixel is:
[0092]
[0093] in, Indicates the first The gray value of the abnormal pixel after correction, Indicates the first The gray value of abnormal pixels, Indicates the first The correction coefficient of abnormal pixels, Represents the ceiling function.
[0094] What needs to be explained is: The value of can be positive or negative, so use right value to be corrected.
[0095] The gray value of each abnormal pixel in the cardiac ultrasound image is replaced with the corrected gray value to obtain a processed cardiac ultrasound image.
[0096] It should be noted that the processed cardiac ultrasound images can help doctors further understand the patient's condition and make more accurate diagnosis results. In this embodiment, when the denominator in the formula is 0, the denominator is set to 1, and this example is used for description to ensure that the formula is valid.
[0097] So far, the present invention is completed.
[0098] In summary, in the embodiment of the present invention, a cardiac ultrasound image and a cardiac ultrasound binary image of the cardiac ultrasound image are obtained; the cardiac ultrasound binary image is divided into a plurality of regions according to the gray value of each pixel on each column in the cardiac ultrasound binary image; the cardiac expression degree of each region is obtained according to the gray value of each pixel on each column in each region and the number of columns in each region; a plurality of cardiac regions are screened out from all regions according to the cardiac expression degree of each region; the region other than the cardiac region in the cardiac ultrasound binary image is equally divided into a plurality of wholes; a plurality of abnormal pixels are screened out from all wholes according to the distance between each whole and each cardiac region, the gray value of each pixel in each whole and the gray value of each pixel in each cardiac region; a correction coefficient of each abnormal pixel is obtained according to the gray value of each abnormal pixel and the gray value of each pixel in the cardiac region closest to each abnormal pixel; a processed cardiac ultrasound image is obtained according to the correction coefficient of each abnormal pixel. So far, the present invention identifies the position of the heart through the characteristic expression of the heart, further analyzes and obtains the distribution expression of the abnormal pixels on the image, and corrects it, thereby improving the image quality of the cardiac ultrasound image.
[0099] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.
Claims
1. An intelligent processing method for cardiac ultrasound images, characterized in that: The method comprises the following steps: Acquire a cardiac ultrasound image and a cardiac ultrasound binary image of the cardiac ultrasound image; According to the gray value of each pixel point on each column in the binary cardiac ultrasound image, the binary cardiac ultrasound image is divided into a plurality of regions; according to the gray value of each pixel point on each column in each region and the number of columns in each region, the cardiac expression degree of each region is obtained; according to the cardiac expression degree of each region, a plurality of cardiac regions are screened out from all regions; The area other than the heart area in the cardiac ultrasound binary image is equally divided into a number of wholes; according to the distance between each whole and each heart area, the gray value of each pixel in each whole and the gray value of each pixel in each heart area, a number of abnormal pixels are screened out from all the wholes; according to the gray value of each abnormal pixel and the gray value of each pixel in the heart area closest to each abnormal pixel, a correction coefficient of each abnormal pixel is obtained; according to the correction coefficient of each abnormal pixel, a processed cardiac ultrasound image is obtained; The step of obtaining the cardiac expression degree of each region according to the gray value of each pixel point on each column in each region and the number of columns in each region includes the following specific steps: Each area from top to bottom The area in the row, from bottom to top The area in the row, from left to right The area in the column, from right to left The areas in the columns are recorded as target areas, where A first quantity threshold is preset; The pixels with continuous adjacent gray values of 1 in each target area form a connected domain, and several connected domains are obtained; When the number of pixels in the connected domain in each target area is greater than or equal to When the target area is recorded as the main target area, A second quantity threshold is preset; The cardiac expressivity of each region is obtained according to the number of main target regions in each region, the grayscale value of each pixel on each column in each region, and the number of columns in each region; According to the number of main target areas in each area, the gray value of each pixel on each column in each area, and the number of columns in each area, the specific calculation formula for obtaining the cardiac expression degree of each area is: in, Indicates The cardiac performance of each region, Indicates The maximum value of the number of pixels between the first pixel with a grayscale value of 0 and the last pixel with a grayscale value of 0 in all columns in the region, Indicates Region The number of pixels with grayscale value 0 in the column, Indicates The number of columns in the region, represents the linear normalization function, For the The number of primary target areas in the region; The method of obtaining the correction coefficient of each abnormal pixel according to the grayscale value of each abnormal pixel and the grayscale value of each pixel in the heart area closest to each abnormal pixel includes the following specific steps: According to the one-to-one correspondence between the cardiac ultrasound image and the pixel points in the cardiac ultrasound binary image, the first The specific calculation formula for the correction coefficient of an abnormal pixel is: in, Indicates The correction coefficient of abnormal pixels, Indicates The overall abnormal performance degree of the abnormal pixel point, Indicates The gray value of the pixel corresponding to each abnormal pixel in the cardiac ultrasound image; Indicates the distance The center point of the heart area closest to the abnormal pixel point corresponds to the gray value of the pixel point in the cardiac ultrasound image; Indicates the distance The grayscale mean of all pixels in the heart area closest to the abnormal pixel corresponding to the pixel in the cardiac ultrasound image, Indicates The abnormal pixel point and the distance The distance between the center points of the heart area closest to the abnormal pixel points, is a linear normalization function.
2. The intelligent processing method of cardiac ultrasound images according to claim 1, characterized in that: The method of dividing the cardiac ultrasound binary image into a plurality of regions according to the grayscale value of each pixel point in each column of the cardiac ultrasound binary image comprises the following specific steps: In the cardiac ultrasound binary image, the column with the largest number of pixels with a gray value of 1 in all columns is used as the region division column, and the cardiac ultrasound binary image is divided into several regions.
3. The intelligent processing method of cardiac ultrasound images according to claim 1, characterized in that: The method of selecting a plurality of cardiac regions from all regions according to the cardiac expression of each region includes the following specific steps: The cardiac expression indexes of all regions are arranged in descending order to obtain a cardiac expression index sequence of all regions, and the regions corresponding to the first two values in the cardiac expression index sequence are recorded as cardiac regions.
4. The intelligent processing method of cardiac ultrasound images according to claim 1, characterized in that: The specific steps of screening out a number of abnormal pixels from all the entities according to the distance between each entity and each heart region, the gray value of each pixel in each entity, and the gray value of each pixel in each heart region are as follows: Jordi The whole is in the In the region, The area is the distance the nearest region to the whole; Jordi The whole is not all in the In the area, calculate the The center point of the whole and the existence of The distance between the center points of each region of the whole will be The area with the shortest distance to the center point of the whole is recorded as the distance the nearest region to the whole; The degree of abnormal performance of each whole is obtained according to the area closest to each whole, the gray value of each pixel in each whole, and the gray value of each pixel in each heart area; The whole with an abnormal performance degree greater than or equal to a preset judgment threshold is recorded as an abnormal whole, and the pixel points in the abnormal whole are recorded as abnormal pixels.
5. The intelligent processing method of cardiac ultrasound images according to claim 4, characterized in that: The specific calculation formula for obtaining the abnormal performance degree of each entity according to the area closest to each entity, the gray value of each pixel in each entity, and the gray value of each pixel in each heart area is: in, Performance The overall degree of abnormal performance, Indicates the distance The cardiac expression of the nearest region of the whole, Indicates The distance between the center point of the whole and the center point of the nearest heart region, Indicates The number of pixels with gray value 0 in the whole Indicates The number of pixels in the whole, Represents the grayscale mean of all pixels in all heart regions; Indicates The grayscale mean of all pixels in the whole is the linear normalization function, represents the absolute value function.
6. The intelligent processing method of cardiac ultrasound images according to claim 1, characterized in that: The specific steps of obtaining the processed cardiac ultrasound image according to the correction coefficient of each abnormal pixel point are as follows: Determine abnormal pixel points in the cardiac ultrasound image and correction coefficients of the abnormal pixel points according to a one-to-one correspondence between the cardiac ultrasound image and the pixel points in the cardiac ultrasound binary image; Obtaining a corrected grayscale value of each abnormal pixel in the cardiac ultrasound image according to a correction coefficient of each abnormal pixel in the cardiac ultrasound image and a grayscale value of each abnormal pixel; The gray value of each abnormal pixel in the cardiac ultrasound image is replaced with the corrected gray value to obtain a processed cardiac ultrasound image.
7. The intelligent processing method of cardiac ultrasound images according to claim 6, characterized in that: The method of obtaining the corrected grayscale value of each abnormal pixel in the cardiac ultrasound image according to the correction coefficient of each abnormal pixel in the cardiac ultrasound image and the grayscale value of each abnormal pixel includes the following specific steps: Calculate the first The correction coefficient of the abnormal pixel point is The product of the grayscale values of the abnormal pixel points is added to the grayscale value of the abnormal pixel point in the cardiac ultrasound image. The sum of the grayscale values of the abnormal pixels is rounded upwards and recorded as the value of the first abnormal pixel in the cardiac ultrasound image. The gray value of the abnormal pixel after correction.
