Image filtering based cardiac ultrasound image optimization enhancement method

Abstract

The application discloses a heart ultrasound image optimization enhancement method based on image filtering and belongs to the technical field of heart ultrasound images. The heart ultrasound image optimization enhancement method based on image filtering comprises the following steps: S1, obtaining a heart ultrasound binary image; S2, obtaining a heart ultrasound image to be processed; S3, highlighting the edge features of a blood vessel wall and myocardial tissue; S4, obtaining an enhanced heart ultrasound image; and S5, obtaining an optimized and enhanced heart ultrasound image. The application solves the problem that the prior art is prone to affecting the overall consistency judgment of heart cavities, valves and other structures, can more intuitively display a heart disease area, reduces the misjudgment of medical staff caused by noise or low contrast, can effectively capture texture information at different levels, enhances the recognition ability of complex heart structures, and further more accurately quantifies myocardial strain rate, ventricular volume and other parameters, thereby providing a reliable basis for evaluating heart function.

Description

Technical Field

[0001] This invention relates to the field of cardiac ultrasound imaging technology, specifically to a method for optimizing and enhancing cardiac ultrasound images based on image filtering. Background Technology

[0002] Cardiac ultrasound is a commonly used non-invasive diagnostic method for assessing the structure and function of the heart and blood vessels. To improve the clarity of ultrasound images and the visualization of the heart and blood vessels, image data optimization and enhancement are necessary. Among these methods, the Canny edge detection algorithm is a commonly used optimization and enhancement method that can accurately extract the boundaries of the heart and blood vessels, segment the edge contours of the heart and blood vessels, and thus enhance the visibility of vascular ultrasound images.

[0003] Chinese Patent Publication No. CN117474775A discloses a method for optimizing and enhancing cardiac ultrasound images, comprising: acquiring a cardiac ultrasound grayscale image; performing image block operation on the cardiac ultrasound grayscale image to obtain several image blocks; classifying all image blocks of the cardiac ultrasound grayscale image, and adaptively cropping the grayscale histogram of each image block to obtain an equalized grayscale histogram, thereby obtaining all equalized image blocks; obtaining all protruding regions of each equalized image block based on the grayscale curve of each equalized image block, obtaining the linear adjustment coefficient of each equalized image block, and enhancing each equalized image block to obtain an enhanced cardiac ultrasound image. This method can effectively reduce the poor image display effect caused by amplified original noise during local histogram equalization enhancement of the image.

[0004] In practical applications, the aforementioned patent performs image segmentation on grayscale cardiac ultrasound images. This segmentation may disrupt the continuity of the target region, resulting in breaks or inconsistencies in the segmentation results at block boundaries, affecting the overall consistency of the judgment of cardiac chambers, valves, and other structures. Therefore, it does not meet the existing requirements. To address this, we propose a cardiac ultrasound image optimization and enhancement method based on image filtering. Summary of the Invention

[0005] The purpose of this invention is to provide a cardiac ultrasound image optimization and enhancement method based on image filtering, which can enhance the contrast between different tissues, avoid the edge blurring problem that may be caused by traditional smoothing filtering, and at the same time, the optimized cardiac ultrasound image can more intuitively display the cardiac lesion area, reduce the misjudgment caused by noise or low contrast by medical staff, effectively capture texture information at different levels, enhance the ability to identify complex cardiac structures, and thus more accurately quantify parameters such as myocardial strain rate and ventricular volume, providing a reliable basis for assessing cardiac function, and solving the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for optimizing and enhancing cardiac ultrasound images based on image filtering, comprising the following steps:

[0007] S1: Obtain the depth map of cardiac ultrasound and obtain a binary image of cardiac ultrasound based on the depth map;

[0008] S2: Perform image filtering on the binarized cardiac ultrasound image to obtain the cardiac ultrasound image to be processed;

[0009] S3: Employing the Laplacian operator to highlight the edge features of the vessel walls and myocardial tissue in the cardiac ultrasound image to be processed. Specifically, this includes: reducing edge artifacts through multi-angle acoustic wave synthesis; calculating the absolute value of the cardiac ultrasound image to be processed using the Laplacian operator, normalizing the calculation results to enhance the edge response of the cardiac ultrasound image to be processed; performing contrast stretching on the cardiac ultrasound image to be processed based on the results of the Laplacian operator to highlight weak edge features, while repairing broken edges in the cardiac ultrasound image to be processed and refining edge lines; superimposing the detected edges of the cardiac ultrasound image to be processed with the original cardiac ultrasound image to obtain the contours of the vessel walls and myocardial tissue of the heart; evaluating the detection effect through edge continuity and contrast indices, and determining whether to adjust the Laplacian parameters based on the detection results.

[0010] S4: Extract texture features by grayscale contrast to enhance the edge contrast between blood vessel walls and myocardial tissue in the cardiac ultrasound image to be processed, thus obtaining an enhanced cardiac ultrasound image;

[0011] S5: Adjust and optimize the quality of the enhanced cardiac ultrasound image to obtain an optimized enhanced cardiac ultrasound image. Specifically, this includes: setting a preset micro-adjustment range threshold; obtaining the dynamic optimal micro-adjustment range based on the preset micro-adjustment range threshold; optimizing the dynamics of the enhanced cardiac ultrasound image based on the optimal micro-adjustment range while suppressing residual speckle noise; performing local adaptive filtering on the key apical region of the optimized cardiac ultrasound image; adjusting the gain and smoothing parameters separately; magnifying the region of interest; and performing secondary edge sharpening processing on the region of interest.

[0012] Preferably, acquiring the depth map of cardiac ultrasound specifically includes:

[0013] Acquire the contour image and high-dimensional image data of the cardiac ultrasound image to be processed;

[0014] First depth images and normal images of multiple different scales are extracted using high-dimensional image data;

[0015] The final depth map is extracted from the normal image and multiple first depth images at different scales.

[0016] Preferably, obtaining the binary image of cardiac ultrasound from the cardiac ultrasound image specifically includes:

[0017] Adjust the gain compensation and transverse gain compensation in cardiac ultrasound images to optimize the signal-to-noise ratio of the ultrasound signal.

[0018] The raw ultrasound signals of cardiac ultrasound images are divided into two categories: blood and tissue, and the boundary points between blood and tissue are identified.

[0019] The endocardial contour is extracted from the tissue, and a binarized image is generated after processing the endocardial contour.

[0020] Preferably, the step of processing the endocardial contour to generate a binary image specifically includes:

[0021] Extract the endocardial contour from the tissue, fill in the local broken areas of the endocardial contour, and connect the discrete boundary points of the local broken areas to form a continuous contour.

[0022] By combining echocardiogram images from different sections, the three-dimensional morphology of the heart structure can be comprehensively determined.

[0023] By utilizing the three-dimensional morphology of the heart structure at different times to form dynamic images, temporal analysis is performed on the dynamic images;

[0024] Based on the time series analysis results, isolated noise points are removed by erosion and dilation operations, jagged edges are smoothed, and the heart chambers are filled to generate a coherent binary mask image.

[0025] The significant edge regions of the binarized mask image are extracted by threshold segmentation to generate a binarized image.

[0026] Preferably, the image filtering processing of the binary cardiac ultrasound image specifically includes:

[0027] Erosion and dilation operations are performed on the binarized image, and morphological filtering is performed on the binarized cardiac ultrasound image.

[0028] Mark all bright pixel regions in the binarized image as independent connected regions and connect the regions;

[0029] Connected regions with an area greater than a threshold are retained. The retained target regions are then processed again before the filtered cardiac ultrasound image is output.

[0030] Preferably, the extraction of texture features through grayscale contrast specifically includes:

[0031] Adjust the dynamic range and gain parameters of the cardiac ultrasound image to be processed, and optimize the grayscale distribution in local areas;

[0032] Based on the vascular orientation of the cardiac ultrasound image to be processed, grayscale compression is performed on the cardiac ultrasound image in both horizontal and vertical directions.

[0033] A gray-level co-occurrence matrix is ​​constructed, and the contrast, entropy, and homogeneity feature maps of the blood vessel wall and myocardium in the cardiac ultrasound image to be processed are extracted from the gray-level co-occurrence matrix.

[0034] The contrast, entropy, and homogeneity feature maps of the blood vessel wall and myocardium are weighted and fused to generate a comprehensive texture feature map.

[0035] Preferably, enhancing the edge contrast between the vessel wall and myocardial tissue in the cardiac ultrasound image to be processed specifically includes:

[0036] Adjust the weight coefficients of contrast, entropy, and homogeneity feature maps to enhance the edge response of the comprehensive texture feature map;

[0037] Decompose and enhance the high-frequency components of the composite texture feature map, and extract the edge direction information of the composite texture feature map.

[0038] The complexity difference of the organizational structure is calculated and quantified based on the high-frequency components of the comprehensive texture feature map and edge direction information.

[0039] Based on the differences in the complexity of the quantified organizational structure, the evolution is preferentially directed towards regions with higher complexity differences to segment the inner membrane boundary;

[0040] An unsharpened mask is applied to the segmented boundary region to enhance the high-frequency components at the edges, making the gray-level jump at the junction of the blood vessel wall and myocardium more significant, thus obtaining the edge contrast enhancement result of the blood vessel wall and myocardial tissue in the cardiac ultrasound image to be processed.

[0041] Preferably, obtaining the dynamically optimal fine-adjustment range based on a preset fine-adjustment range threshold specifically includes:

[0042] When the fine-tuning range is greater than or equal to the preset fine-tuning range threshold, the target limiting parameter is taken as the optimal limiting parameter.

[0043] When the fine-tuning range is less than the preset fine-tuning range threshold, the target limit parameter is added to the preset parameter step size to obtain a new target limit parameter, and the fine-tuning range of the new target limit parameter is obtained.

[0044] Repeat the judgment operation on the new target constraint parameters until the optimal constraint parameters are obtained, and stop the iteration to obtain the dynamic optimal micro-adjustment range of cardiac ultrasound image quality.

[0045] Compared with the prior art, the beneficial effects of the present invention are:

[0046] This invention effectively reduces speckle noise and random noise in echocardiogram images by filtering them, thereby enhancing the clarity of the boundaries of structures such as the myocardial wall and valves. This helps in identifying minute lesions. While suppressing high-frequency noise, it preserves key anatomical details. Adjusting the dynamic range parameter expands the grayscale information distribution, enhances the contrast between different tissues, and makes features such as blood flow status and myocardial texture easier to identify. It avoids the edge blurring problem that may occur with traditional smoothing filtering. At the same time, the optimized echocardiogram image can more intuitively display the cardiac lesion area, reducing misjudgments by medical staff due to noise or low contrast. By adjusting the dynamic range, it can more clearly display the boundaries and subtle structural differences of myocardial tissue, effectively capture texture information at different levels, enhance the ability to identify complex cardiac structures, and thus more accurately quantify parameters such as myocardial strain rate and ventricular volume, providing a reliable basis for assessing cardiac function. Attached Figure Description

[0047] Figure 1 This is a flowchart of the cardiac ultrasound image optimization and enhancement method based on image filtering according to the present invention;

[0048] Figure 2 This is a schematic diagram illustrating the adjustment and optimization of cardiac ultrasound image quality according to the present invention. Detailed Implementation

[0049] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0050] To address the issue that existing techniques for segmenting grayscale echocardiogram images can disrupt the continuity of the target region, leading to breaks or inconsistencies at block boundaries and affecting the overall consistency of assessments of cardiac chambers, valves, and other structures, please refer to [the relevant documentation / reference]. Figures 1-2 This embodiment provides the following technical solution:

[0051] A method for optimizing and enhancing cardiac ultrasound images based on image filtering includes the following steps:

[0052] S1: Obtain the depth map of cardiac ultrasound and obtain a binary image of cardiac ultrasound based on the depth map;

[0053] S2: Perform image filtering on the binarized cardiac ultrasound image to obtain the cardiac ultrasound image to be processed;

[0054] S3: The Laplacian operator is used to highlight the edge features of the blood vessel wall and myocardial tissue in the cardiac ultrasound image to be processed. Combined with spatial composite imaging technology, multi-angle acoustic wave synthesis reduces edge artifacts.

[0055] S4: Extract texture features by grayscale contrast to enhance the edge contrast between blood vessel walls and myocardial tissue in the cardiac ultrasound image to be processed, thus obtaining an enhanced cardiac ultrasound image;

[0056] S5: Adjust and optimize the quality of the enhanced cardiac ultrasound image to obtain an optimized enhanced cardiac ultrasound image.

[0057] Obtaining depth maps from cardiac ultrasound, specifically including:

[0058] Contour extraction is performed on the cardiac ultrasound image to be processed to obtain the contour image of the cardiac ultrasound image to be processed;

[0059] Extract first depth images at multiple different scales from high-dimensional image data, and extract normal images from high-dimensional image data and multiple first depth images at different scales;

[0060] Based on the normal image and multiple first depth images at different scales, a second depth image is extracted, which becomes the final depth map.

[0061] Obtaining a binary image of cardiac ultrasound from the cardiac ultrasound image specifically includes:

[0062] Adjust the gain compensation and transverse gain compensation in cardiac ultrasound images to optimize the signal-to-noise ratio of the ultrasound signal.

[0063] The raw ultrasound signals of cardiac ultrasound images are divided into two categories: blood and tissue, and the boundary points between blood and tissue are identified.

[0064] The endocardial contour is extracted from the tissue, and a binarized image is generated after processing the endocardial contour.

[0065] After processing the endocardial contour, a binarized image is generated, specifically including:

[0066] Extract the endocardial contour from the tissue, fill in the local broken areas of the endocardial contour, and connect the discrete boundary points of the local broken areas to form a continuous contour.

[0067] By combining echocardiogram images from different sections, the three-dimensional morphology of the heart structure can be comprehensively determined.

[0068] Dynamic images are generated by using the three-dimensional morphology of the heart structure at different times. Temporal analysis is performed on the dynamic images to ensure that the binarized contour is synchronized with the cardiac cycle.

[0069] Based on the time series analysis results, isolated noise points are removed by erosion and dilation operations, jagged edges are smoothed, and the heart chambers are filled to generate a coherent binary mask image.

[0070] The significant edge regions of the binarized mask image are extracted by threshold segmentation to generate a binarized image.

[0071] Image filtering processing is performed on the binarized cardiac ultrasound images, specifically including:

[0072] The erosion operation is performed on the binarized image using structuring elements to eliminate small noise points and isolated bright spots. The eroded image is then subjected to dilation to restore the original shape of the target region while avoiding excessive shrinkage.

[0073] The image is traversed by an 8-neighborhood search algorithm. All bright pixel regions in the binarized image are marked as independent connected regions. The area and perimeter of each connected region are calculated to provide a basis for subsequent screening.

[0074] Based on the characteristics of the heart structure, a minimum area threshold is set to filter noise, retaining only connected regions with an area greater than the threshold and setting the rest of the region as background.

[0075] The target area is expanded again to smooth the jagged edges and improve the continuity of the shape. If there are holes in the target area, the holes are filled.

[0076] The Laplacian operator is used to highlight the edge features of blood vessel walls and myocardial tissue in the cardiac ultrasound images to be processed, specifically including:

[0077] The absolute value of the cardiac ultrasound image to be processed is calculated using the Laplacian operator, and the calculation result is normalized to enhance the edge response of the cardiac ultrasound image to be processed.

[0078] The Laplacian operator results are used to stretch the contrast of the cardiac ultrasound image to be processed, highlighting weak edge features. At the same time, the broken edges of the cardiac ultrasound image to be processed are repaired and the edge lines are refined.

[0079] The detected edges of the cardiac ultrasound image to be processed are superimposed on the original cardiac ultrasound image to obtain the outline of the blood vessel wall and myocardial tissue of the heart.

[0080] The detection performance is evaluated using edge continuity and contrast metrics, and the Laplacian parameters are adjusted based on the results.

[0081] Texture features are extracted through grayscale contrast, specifically including:

[0082] The dynamic range and gain parameters of the cardiac ultrasound image to be processed are adjusted, and the gray-level distribution of the local area is optimized to highlight the gray-level difference between the blood vessel wall and myocardial tissue. The local contrast is enhanced by reducing the dynamic range, while avoiding noise interference caused by excessive gain and reducing the influence of speckle noise, which is common in ultrasound images.

[0083] To optimize the processing of cardiac ultrasound images, grayscale compression is performed on the images in both horizontal and vertical directions to reduce computational load while preserving effective texture feature information, based on the vessel orientation.

[0084] Enhance the gray-level gradient at the myocardial-blood interface using histogram equalization.

[0085] A gray-level co-occurrence matrix is ​​constructed, and the contrast, entropy, and homogeneity feature maps of the blood vessel wall and myocardium in the cardiac ultrasound image to be processed are extracted from the gray-level co-occurrence matrix.

[0086] Contrast is used to reflect edge sharpness; high-contrast areas correspond to the boundary between the blood vessel wall and the myocardium. Entropy is used to quantify texture complexity and to distinguish between homogeneous myocardial tissue and the heterogeneous structure of the blood vessel wall. Homogeneity is used to identify the differences between smooth regions (such as the myocardium) and rough regions (such as the inner wall of a blood vessel).

[0087] The contrast, entropy, and homogeneity feature maps of the blood vessel wall and myocardium are weighted and fused to generate a comprehensive texture feature map.

[0088] Enhancing the edge contrast between blood vessel walls and myocardial tissue in the cardiac ultrasound image to be processed, specifically including:

[0089] Adjust the contrast, entropy, and homogeneity feature map weights (e.g., set the contrast weight to 0.6 and the entropy weight to 0.3) to enhance the edge response of the comprehensive texture feature map;

[0090] Decompose and enhance the high-frequency components of the composite texture feature map, and extract the edge direction information of the composite texture feature map.

[0091] The complexity difference of the organizational structure is calculated and quantified based on the high-frequency components of the comprehensive texture feature map and edge direction information.

[0092] Based on the differences in the complexity of the quantified organizational structure, the evolution is preferentially directed towards regions with higher complexity differences to segment the inner membrane boundary;

[0093] An unsharpened mask is applied to the segmented boundary region to enhance the high-frequency components at the edges, making the gray-level jump at the junction of the blood vessel wall and myocardium more significant, thus obtaining the edge contrast enhancement result of the blood vessel wall and myocardial tissue in the cardiac ultrasound image to be processed.

[0094] Adjustments and optimizations were made to the quality of the enhanced cardiac ultrasound images, specifically including:

[0095] A preset fine-tuning range threshold is used to obtain the dynamic optimal fine-tuning range.

[0096] The dynamics of enhanced cardiac ultrasound images are optimized based on the optimal micro-adjustment range, while suppressing residual speckle noise. The micro-adjustment range (DynRange) is adjusted to 50-70dB to avoid image fogging and improve tissue boundary clarity.

[0097] Local adaptive filtering is applied to the key region of the apex in the optimized echocardiogram image. The gain and smoothing parameters are adjusted separately to magnify the region of interest and perform secondary edge sharpening on the region of interest.

[0098] The dynamic optimal fine-tuning range is obtained based on a preset fine-tuning range threshold, specifically including:

[0099] When the fine-tuning range is greater than or equal to the preset fine-tuning range threshold, the target limiting parameter is taken as the optimal limiting parameter.

[0100] When the fine-tuning range is less than the preset fine-tuning range threshold, the target limit parameter is added to the preset parameter step size to obtain a new target limit parameter, and the fine-tuning range of the new target limit parameter is obtained.

[0101] Repeat the judgment operation on the new target constraint parameters until the optimal constraint parameters are obtained, and stop the iteration to obtain the dynamic optimal micro-adjustment range of cardiac ultrasound image quality.

[0102] In summary, the cardiac ultrasound image optimization and enhancement method based on image filtering of this invention can effectively reduce speckle noise and random noise in cardiac ultrasound images by filtering them. For example, Gaussian filtering preserves edge information and smooths the background through weighted averaging, improving the signal-to-noise ratio and thus enhancing the boundary clarity of structures such as the myocardial wall and valves, which helps in the identification of minute lesions. It also preserves key anatomical details while suppressing high-frequency noise. Adjusting the dynamic range parameter expands the grayscale information distribution, enhances the contrast between different tissues, and makes features such as blood flow status and myocardial texture easier to identify, reducing artifacts caused by cardiac motion. The filtering technique can effectively suppress speckle noise and artifacts in ultrasound images, and the optimization and enhancement process further reduces interference from irrelevant signals, making the contrast between cardiac tissue and background clearer. Filtering optimizes and enhances echocardiogram images, highlighting the boundaries of structures such as the myocardium and valves, reducing blurring caused by noise, and preserving details of subtle pathological features. Through grayscale adjustment and histogram equalization, the dynamic range of the image can be expanded, adaptively protecting the morphological features of the heart structure and avoiding edge blurring problems that may occur with traditional smoothing filtering. The optimized echocardiogram image can more intuitively display areas of cardiac lesions, reducing misjudgments by medical staff due to noise or low contrast. By adjusting the dynamic range, the boundaries and subtle structural differences of myocardial tissue can be displayed more clearly, effectively capturing texture information at different levels, enhancing the ability to identify complex cardiac structures, and thus more accurately quantifying parameters such as myocardial strain rate and ventricular volume, providing a reliable basis for assessing cardiac function.

[0103] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0104] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention.

Claims

1. A method for cardiac ultrasound image optimization enhancement based on image filtering, characterized in that, The method comprises the following steps: S1: obtaining a depth map of the cardiac ultrasound, and obtaining a binary image of the cardiac ultrasound according to the depth map of the cardiac ultrasound; S2: performing image filtering processing on the binary image of the cardiac ultrasound to obtain a to-be-processed cardiac ultrasound image; S3: using a Laplacian operator to highlight the edge features of the blood vessel wall and myocardial tissue in the to-be-processed cardiac ultrasound image, specifically comprising: reducing edge artifacts by multi-angle acoustic wave synthesis; calculating the absolute value of the to-be-processed cardiac ultrasound image by using the Laplacian operator, performing normalization processing on the calculation result, and enhancing the edge response of the to-be-processed cardiac ultrasound image; performing contrast stretching on the result of the Laplacian operator to highlight weak edge features, repair the connected broken edges of the to-be-processed cardiac ultrasound image, and refine the edge lines; superimposing the detected edges of the to-be-processed cardiac ultrasound image on the original image of the cardiac ultrasound to obtain the outline of the blood vessel wall and myocardial tissue of the heart; evaluating the detection effect through edge continuity and contrast index, and determining whether to adjust the Laplacian parameters according to the detection result; S4: extracting texture features through gray contrast to enhance the edge contrast of the blood vessel wall and myocardial tissue in the to-be-processed cardiac ultrasound image, and obtaining an enhanced cardiac ultrasound image; S5: adjusting and optimizing the quality of the enhanced cardiac ultrasound image to obtain an optimized enhanced cardiac ultrasound image, specifically comprising: presetting a micro-adjustment range threshold, obtaining a dynamic optimal micro-adjustment range according to the preset micro-adjustment range threshold; optimizing the dynamics of the enhanced cardiac ultrasound image according to the optimal micro-adjustment range, while suppressing residual speckle noise; performing local adaptive filtering on the key area of the apex of the optimized cardiac ultrasound image, separately adjusting the gain and smoothing parameters, amplifying the region of interest, and performing secondary edge sharpening processing on the region of interest.

2. The image filter based cardiac ultrasound image optimization enhancement method of claim 1, wherein: The depth map of the cardiac ultrasound is obtained, specifically comprising: obtaining a contour image and high-dimensional image data of the to-be-processed cardiac ultrasound image; extracting a plurality of first depth images and normal images of different scales using the high-dimensional image data; extracting a final depth map according to the normal image and the plurality of first depth images of different scales.

3. The image filter based cardiac ultrasound image optimization enhancement method of claim 1, wherein: The binary image of the cardiac ultrasound is obtained according to the cardiac ultrasound image, specifically comprising: adjusting the gain compensation and lateral gain compensation in the cardiac ultrasound image to optimize the signal-to-noise ratio of the ultrasound signal of the cardiac ultrasound image; dividing the original ultrasound signal of the cardiac ultrasound image into two categories of blood and tissue, and identifying the critical points of the blood and tissue; extracting the endocardial contour in the tissue, and generating a binary image after processing the endocardial contour.

4. The image filter based cardiac ultrasound image optimization enhancement method of claim 3, wherein: The binary image is generated after processing the endocardial contour, specifically comprising: extracting the endocardial contour in the tissue, filling the local broken area of the endocardial contour, connecting the discrete boundary points of the local broken area, and forming a continuous contour; comprehensively judging the three-dimensional morphology of the cardiac structure in combination with cardiac ultrasound images of different sections; forming a dynamic image using the three-dimensional morphology of the cardiac structure at different times, and performing time sequence analysis on the dynamic image; According to the time sequence analysis result, isolated noise points are removed through erosion and expansion operations, edge sawtooth is smoothed, and the heart cavity is region filled to generate a coherent binary mask image; A significant edge region of the binary mask image is extracted through threshold segmentation to generate a binary image.

5. The image filter based cardiac ultrasound image optimization enhancement method of claim 1, wherein: The image filtering processing of the heart ultrasound binary image specifically includes: The binary image is subjected to erosion operation and expansion operation, and the heart ultrasound binary image is subjected to morphological filtering; All bright pixel regions in the binary image are marked as independent connected regions, and the regions are connected; Connected regions with an area greater than a threshold value are retained, and the target regions retained are subjected to processing again to output a filtered heart ultrasound image to be processed.

6. The image filter based cardiac ultrasound image optimization enhancement method of claim 1, wherein: The texture feature is extracted through gray scale contrast, specifically including: The dynamic range and gain parameters of the heart ultrasound image to be processed are adjusted, and the gray scale distribution of a local region is optimized; The heart ultrasound image to be processed is subjected to gray scale compression in the horizontal and vertical directions according to the blood vessel direction of the heart ultrasound image to be processed; A gray scale co-occurrence matrix is constructed, and the contrast, entropy and homogeneity feature maps of the blood vessel wall and myocardium in the heart ultrasound image to be processed are extracted from the gray scale co-occurrence matrix; The contrast, entropy and homogeneity feature maps of the blood vessel wall and myocardium are fused to generate a comprehensive texture feature map.

7. The image filter based cardiac ultrasound image optimization enhancement method of claim 1, wherein: The edge contrast of the blood vessel wall and myocardial tissue in the heart ultrasound image to be processed is enhanced, specifically including: The contrast, entropy and homogeneity feature map weight coefficients are adjusted to strengthen the edge response of the comprehensive texture feature map; The high-frequency component of the enhanced comprehensive texture feature map is decomposed to extract the edge direction information of the enhanced comprehensive texture feature map; The complexity difference of the quantized tissue structure is calculated according to the high-frequency component of the comprehensive texture feature map and the edge direction information; According to the complexity difference of the quantized tissue structure, the intimal boundary is preferentially evolved to the region with high complexity difference; The segmented boundary region is subjected to non-sharpening mask to enhance the high-frequency component of the edge, so that the gray scale jump at the junction of the blood vessel wall and myocardium is more significant, and the edge contrast enhancement result of the blood vessel wall and myocardial tissue in the heart ultrasound image to be processed is obtained.

8. The image filter based cardiac ultrasound image optimization enhancement method of claim 5, wherein: The dynamic optimal micro-adjustment range is obtained according to the preset micro-adjustment range threshold, specifically including: When the micro-adjustment range is greater than or equal to the preset micro-adjustment range threshold, the target limit parameter is taken as the optimal limit parameter; When the micro-adjustment range is less than the preset micro-adjustment range threshold, the target limit parameter is added to the preset parameter step to obtain a new target limit parameter, and the micro-adjustment range of the new target limit parameter is obtained; The judgment operation on the new target limit parameter is repeated until the optimal limit parameter is obtained to stop iteration, and the dynamic optimal micro-adjustment range of the heart ultrasound image quality is obtained.

Images