Intelligent enhancement method for low-dose clinical medicine ultrasound contrast

Abstract

The invention relates to the technical field of ultrasonic image enhancement, in particular to a low-dose clinical medicine ultrasonic contrast intelligent enhancement method, which comprises the following steps: acquiring a blood vessel ultrasonic contrast image of a patient under a low-dose contrast agent; acquiring the extension degree of each pixel in the vascular ultrasound contrast image from the normal vascular reference area; acquiring the dispersion degree of a preset local area where each pixel in the vascular ultrasound contrast image is located; acquiring a plurality of blood vessel regions with fine structures in the blood vessel ultrasound contrast image; obtaining the final degree that a blood vessel width region corresponding to each pixel in the skeleton line of the blood vessel region with the fine structure belongs to a focus; and two types of pixels to be enhanced in the vascular ultrasound contrast image are acquired and enhanced, and an enhanced image of the vascular ultrasound contrast image is obtained. According to the method, the contrast ratio of the blood vessel region and the surrounding tissues is improved, so that the blood vessel region with a fine structure can be better identified, and the imaging effect of the ultrasonic contrast image is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of ultrasonic image enhancement, and in particular to a low-dose clinical medical ultrasonic contrast intelligent enhancement method. Background Art

[0002] Ultrasound contrast imaging, also known as acoustic contrast imaging, is a technique that uses contrast agents to enhance backscattered echoes, significantly improving the resolution, sensitivity and specificity of ultrasound diagnosis. This technique combines ultrasonic imaging and microbubble contrast agents to more clearly show the blood flow of organs and tissues, as well as detect vascular problems or tumors and other lesions. Compared with other imaging techniques, ultrasound contrast imaging has the advantages of simple operation, no radiation, no trauma, and no pain, so it is highly favored in clinical applications.

[0003] In clinical medicine, when low-dose contrast agents are used to obtain ultrasound contrast-enhanced images of patients, factors such as the patient's condition and tissue type make it impossible for the contrast agent to be completely evenly distributed in different locations of the blood vessels, resulting in insufficient signal intensity at the fine structures of the blood vessels and insufficient contrast between the blood vessels and surrounding tissues, which affects the identification of fine-structured vascular areas. The ultrasound contrast-enhanced imaging effect is poor, which is not conducive to the detection of vascular lesions in patients. Therefore, it is very important to intelligently enhance the patient's ultrasound contrast-enhanced images. Summary of the invention

[0004] In order to solve the above problems, the present invention provides a low-dose clinical medical ultrasound contrast intelligent enhancement method.

[0005] A low-dose clinical medical ultrasound contrast intelligent enhancement method of the present invention adopts the following technical solution: An embodiment of the present invention provides a low-dose clinical medical ultrasound contrast intelligent enhancement method, the method comprising the following steps: Obtain vascular ultrasound contrast images of patients under low-dose contrast agents; According to the grayscale expression of the blood vessel area in the vascular ultrasound contrast image, a normal blood vessel reference area is obtained; according to the distance extension of the pixel from the normal blood vessel reference area in the vascular ultrasound contrast image, the concentrated expression of the grayscale value within the preset neighborhood range of the pixel and the brightness expression within the preset neighborhood range, the extension degree of each pixel from the normal blood vessel reference area in the vascular ultrasound contrast image is obtained; Obtain high-brightness pixels in a preset local area of ​​each pixel in the vascular ultrasound contrast imaging; obtain the degree of diffusion of the preset local area where each pixel in the vascular ultrasound contrast imaging is located according to the proportion of high-brightness pixels in the preset local area, the discrete distribution of high-brightness pixels and the degree of confusion of pixel grayscale in the preset local area; According to the extension degree and the diffusion degree, the possibility of each pixel in the vascular ultrasound contrast image belonging to the vascular region with fine structure is obtained; according to the possibility that the pixel belongs to the vascular region with fine structure, several vascular regions with fine structure are obtained in the vascular ultrasound contrast image; Obtain the blood vessel width area corresponding to each pixel point on the skeleton line of the finely structured blood vessel area; obtain the final degree to which the blood vessel width area corresponding to each pixel point on the skeleton line of each finely structured blood vessel area belongs to the lesion according to the grayscale expression of the blood vessel width area corresponding to the pixel point on the skeleton line, the distance of the blood vessel width area from both ends of the finely structured blood vessel area, and the possibility that the pixel belongs to the finely structured blood vessel area; According to the final degree of the lesion of the vascular width area corresponding to each pixel in the skeleton line of the finely structured vascular area, two types of pixels to be enhanced in the vascular ultrasound contrast image are obtained; different methods are used to enhance the two types of pixels to be enhanced in the vascular ultrasound contrast image, and an enhanced image of the patient's vascular ultrasound contrast image under low-dose contrast agent is obtained.

[0006] Furthermore, the method of obtaining the extension degree of each pixel in the vascular ultrasound contrast image from the normal blood vessel reference area according to the distance extension of the pixel in the vascular ultrasound contrast image from the normal blood vessel reference area, the concentrated expression of the gray value in the preset neighborhood range of the pixel, and the brightness expression in the preset neighborhood range includes the following specific steps: Vascular ultrasound imaging The Euclidean distance of the pixel from the center of the normal blood vessel reference area is taken as the The distance extension feature of the pixel from the normal blood vessel reference area; The absolute difference between the grayscale value of each pixel in the preset neighborhood range and the grayscale mean value of all pixels in the preset neighborhood range is recorded as the initial grayscale concentration degree. The inverse proportional value of the accumulated value of the initial grayscale concentration of all pixels within the preset neighborhood of the pixel is used as the The concentrated expression characteristics of the grayscale value within the preset neighborhood range of pixels; The grayscale mean of all pixels within the preset neighborhood of the pixel is taken as the The brightness performance characteristics within the preset neighborhood range of pixels; the distance extension characteristics, the concentration performance characteristics, and the brightness performance characteristics are fused and normalized to obtain the first pixel in the vascular ultrasound angiography image. The extension degree of each pixel from the normal blood vessel reference area; the distance extension feature, the concentration performance feature, and the brightness performance feature are all proportional to the extension degree.

[0007] Furthermore, the method of obtaining the diffusion degree of the preset local area where each pixel in the vascular ultrasound contrast imaging is located according to the proportion of high-brightness pixels in the preset local area, the discrete distribution of high-brightness pixels and the degree of confusion of pixel grayscale in the preset local area includes the following specific steps: Vascular ultrasound imaging The number of high brightness pixels in the preset local area of ​​pixels is The ratio of the number of all pixels in the preset local area of ​​pixels is taken as the first The ratio characteristics of high brightness pixels in the preset local area of ​​pixels; The variance of the Euclidean distance between high-brightness pixels in the preset local area of ​​pixels is used as the The discrete distribution characteristics between high-brightness pixels in the preset local area of ​​pixels; The information entropy of different grayscale values ​​in the preset local area of ​​pixels is used as the The chaos degree feature of the pixel grayscale in the preset local area of ​​pixels is obtained; the proportion feature, the discrete distribution feature, and the chaos degree feature are fused and normalized to obtain the first The diffusion degree of the preset local area where the pixels are located; the proportion feature is inversely proportional to the diffusion degree, and the discrete distribution feature and the chaos degree feature are directly proportional to the diffusion degree.

[0008] Furthermore, the possibility of each pixel in the vascular ultrasound contrast imaging belonging to a vascular region with a fine structure is obtained according to the extension degree and the diffusion degree, and the specific steps include the following: The normalized value of the product of the extension degree and the diffusion degree is used as the first The probability that a pixel belongs to a vascular region with subtle structure.

[0009] Furthermore, the method of obtaining several vascular regions with fine structures in the vascular ultrasound contrast imaging according to the possibility that the pixel belongs to the vascular region with fine structures includes the following specific steps: Pixels whose probability of belonging to the subtle vascular area is greater than the preset probability threshold are taken as pixels to be analyzed, and DBSCAN density clustering is performed on all pixels to be analyzed in the vascular ultrasound contrast images to obtain several clusters. The convex hull detection algorithm is used to perform convex hull detection on each cluster, and the minimum convex hull area corresponding to each cluster is taken as the subtle vascular area in the vascular ultrasound contrast images.

[0010] Furthermore, the specific method for obtaining the blood vessel width area corresponding to each pixel point on the skeleton line of the blood vessel area with fine structure is as follows: Any fine-structured vascular region in the vascular ultrasound angiography image is recorded as the current fine-structured vascular region; the vascular width region corresponding to each pixel point on the skeleton line of the current fine-structured vascular region is obtained, and the vascular width region is specifically a line segment region formed by the intersection of the pixel points on the skeleton line and the edges on both sides of the current fine-structured vascular region.

[0011] Furthermore, the method of obtaining the final degree to which the blood vessel width area corresponding to each pixel in the skeleton line of each fine-structured blood vessel area belongs to the lesion according to the grayscale expression of the blood vessel width area corresponding to the pixel point on the skeleton line, the distance between the blood vessel width area and the two ends of the fine-structured blood vessel area, and the possibility that the pixel belongs to the fine-structured blood vessel area, includes the following specific steps: In the formula, The skeleton line of the current fine-structured blood vessel region The average gray value of the blood vessel width area corresponding to pixels; The skeleton line of the current fine-structured blood vessel region The number of pixels within a preset neighborhood of pixels; is the lesion parameter, when hour, ,when hour, , is the preset first value, is the preset second value, ; The skeleton line of the current fine-structured blood vessel region The preset neighboring range of pixels The average gray value of the blood vessel width area corresponding to pixels; It is the average of the average grayscale values ​​of the blood vessel width area corresponding to all pixels in the skeleton line of the current fine-structured blood vessel area; The first The distance of pixels from one end of the current fine-structured blood vessel area; The first The distance of pixels from the other end of the current fine-structured blood vessel area; To take the absolute value; The skeleton line of the current fine-structured blood vessel region The blood vessel width area corresponding to pixels belongs to the initial extent of the lesion; According to the possibility that the pixel belongs to the fine-structured vascular region and the initial degree to which the vascular width region corresponding to the pixel belongs to the lesion, the final degree to which the vascular width region corresponding to each pixel in the skeleton line of each fine-structured vascular region belongs to the lesion is obtained.

[0012] Furthermore, the method of obtaining the final degree to which the blood vessel width area corresponding to each pixel in the skeleton line of each blood vessel area with fine structure belongs to the lesion according to the possibility that the pixel belongs to the blood vessel area with fine structure and the initial degree to which the blood vessel width area corresponding to the pixel belongs to the lesion comprises the following specific steps: In the formula, The skeleton line of the current fine-structured blood vessel region The average probability that all pixels in the blood vessel width area corresponding to pixels belong to the blood vessel area with fine structure; To prevent hyperparameters with denominators equal to 0; is a linear normalization function; The skeleton line of the current fine-structured blood vessel region The blood vessel width area corresponding to pixels belongs to the final extent of the lesion.

[0013] Furthermore, the method of obtaining two types of pixels to be enhanced in the vascular ultrasound contrast imaging according to the final extent to which the vascular width area corresponding to each pixel in the skeleton line of the finely structured vascular area belongs to the lesion includes the following specific steps: For all the fine-structured vascular regions in the vascular ultrasound contrast imaging, the pixels in the vascular width region whose final degree is greater than the preset lesion threshold are regarded as lesion-type pixels; otherwise, they are regarded as fine-structured vascular-type pixels.

[0014] Furthermore, the two types of pixels to be enhanced in the vascular ultrasound contrast imaging are enhanced by different methods to obtain an enhanced image of the patient's vascular ultrasound contrast imaging under low-dose contrast agent, including the following specific steps: The grayscale mean of the background and the grayscale mean of the normal blood vessel reference area in the vascular ultrasound contrast-enhanced image are obtained, wherein the grayscale mean of the background is a preset parameter, and the grayscale value of the blood vessel-like pixels with fine structure in the vascular ultrasound contrast-enhanced image is set to the pixel mean of the normal blood vessel reference area, and the grayscale value of the lesion-like pixels in the vascular ultrasound contrast-enhanced image is set to the absolute difference between the grayscale mean of the background and the grayscale mean of the normal blood vessel reference area. times, is a preset third value, and an enhanced image of the patient's vascular ultrasound contrast imaging under a low-dose contrast agent is obtained.

[0015] The beneficial effect of the technical solution of the present invention is as follows: the present invention determines the vascular area with fine structure in the vascular ultrasound contrast image by analyzing the extension degree of pixels from the normal vascular reference area and the diffusion degree of the preset local area where the pixels are located in the vascular ultrasound contrast image, and analyzes the lesions and vascular areas in the vascular area with fine structure to obtain two types of pixels to be enhanced in the vascular ultrasound contrast image, and by adopting different methods to enhance the two types of pixels to be enhanced in the vascular ultrasound contrast image, an enhanced image of the vascular ultrasound contrast image of the patient under low-dose contrast agent is obtained, the contrast between the vascular area and the surrounding tissue is improved, and the lesions are also enhanced and analyzed, so that the vascular area with fine structure can be better identified, the imaging effect of the ultrasound contrast image is improved, and the intelligent enhancement of the ultrasound contrast image is realized. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] 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.

[0017] Figure 1 A flowchart of the steps of a low-dose clinical medical ultrasound contrast intelligent enhancement method provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] 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 low-dose clinical medical ultrasound contrast intelligent enhancement method proposed by the present invention, its specific implementation method, structure, characteristics and effects, in combination 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.

[0019] 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.

[0020] The specific scheme of the low-dose clinical medical ultrasound contrast intelligent enhancement method provided by the present invention is described in detail below with reference to the accompanying drawings.

[0021] See also Figure 1 , which shows a flowchart of a method for intelligent enhancement of low-dose clinical medical ultrasound contrast imaging provided by an embodiment of the present invention, the method comprising the following steps: Step S001: Acquire a vascular ultrasound contrast image of a patient under low-dose contrast agent.

[0022] It should be noted that the main purpose of this embodiment is to perform intelligent enhancement on the vascular area with fine structure in the vascular ultrasound contrast image. Before starting the analysis, the vascular ultrasound contrast image of the patient under low-dose contrast agent is first obtained.

[0023] Specifically, the patient's vascular ultrasound contrast imaging under low-dose contrast agent is obtained as follows: During the examination, the medical staff first establishes an intravenous channel (forearm vein) for the patient, and the contrast agent is injected at a uniform speed through the established intravenous channel. The injection speed and dosage are determined according to the purpose of the test and the type of contrast agent. This implementation will not go into details. After the contrast agent is injected, the medical staff obtains the patient's vascular ultrasound angiography image under low-dose contrast agent through an ultrasonic image diagnostic device.

[0024] It should be noted that before obtaining the image, the doctor or nurse needs to ensure that the patient is suitable for ultrasound contrast examination, and remind the patient to wear comfortable clothes before the test and not eat 6-8 hours before the test, because food residue or gas will interfere with the test results. The obtained vascular ultrasound contrast image is a grayscale image.

[0025] At this point, the patient's vascular ultrasound angiography images under low-dose contrast agent are obtained.

[0026] Step S002, obtaining a normal blood vessel reference area according to the grayscale expression of the blood vessel area in the vascular ultrasound angiography image; obtaining the extension degree of each pixel in the vascular ultrasound angiography image from the normal blood vessel reference area according to the distance extension of the pixel in the vascular ultrasound angiography image from the normal blood vessel reference area, the concentrated expression of the grayscale value within the preset neighborhood range of the pixel, and the brightness expression within the preset neighborhood range.

[0027] It should be noted that in order to improve the imaging effect of blood vessels in some fine structures in ultrasound contrast-enhanced images, it is first necessary to obtain the fine-structured blood vessel areas in ultrasound contrast-enhanced images. The area around normal blood vessels is relatively smooth. After microbubbles (microbubble contrast agents) are injected into the patient's blood vessels, they will circulate in the body with the blood flow. When the ultrasound beam irradiates the microbubbles in the contrast agent, the microbubbles will produce a strong echo signal, which is much stronger than other tissues around the vascular channel. Therefore, on the ultrasound contrast-enhanced image, the blood vessel area will appear as a high-brightness area, showing a neat and continuous bright line; while for the fine-structured blood vessel area, its size is small, and the ultrasonic resolution ability of the equipment is limited. It appears as a diffuse highlight in the ultrasound contrast-enhanced image, but these fine-structured blood vessel areas are connected to the surrounding normal blood vessel areas as a whole, forming a blood vessel network with good extensibility, and are branches of some normal blood vessel areas. Therefore, consider analyzing the extension degree of the current pixel position from the normal blood vessel and the diffusion degree of the local area where the current pixel is located, and obtain the possibility that the current pixel belongs to the fine-structured blood vessel area, and then obtain the fine-structured blood vessel area. First, obtain the normal blood vessel reference area.

[0028] Specifically, according to the grayscale expression of the blood vessel area in the vascular ultrasound contrast image, the normal blood vessel reference area is obtained, as follows: In the vascular ultrasound angiography image, any pixel with the highest grayscale value is selected as the initial point for region growing. The growth criterion of region growing is: the pixels within the eight neighborhoods of the initial point whose grayscale absolute difference with the initial point is less than The neighboring pixels of are included into the region of the initial point until no more neighboring pixels meet the growth criterion, and the region growth is stopped to obtain the growth region of the initial point. is the preset first threshold value. Describe; obtain the growth area corresponding to all pixels with the highest gray value in the vascular ultrasound contrast image, and use the area with the largest growth area as the normal blood vessel reference area.

[0029] It should be noted that the normal blood vessel reference area is obtained as mentioned above, and then the extension degree of each pixel in the vascular ultrasound contrast image from the normal blood vessel reference area is obtained by analyzing the distance extension of the pixel from the normal blood vessel reference area in the ultrasound contrast image, the concentration expression of the grayscale value within the preset neighborhood range of the pixel, and the brightness expression within the preset neighborhood range.

[0030] Preferably, in one embodiment of the present invention, the extension degree of each pixel in the vascular ultrasound contrast image from the normal blood vessel reference area is obtained according to the distance extension of the pixel in the vascular ultrasound contrast image from the normal blood vessel reference area, the concentrated expression of the gray value within the preset neighborhood range of the pixel, and the brightness expression within the preset neighborhood range. The specific method is as follows: Vascular ultrasound imaging The Euclidean distance of the pixel from the center of the normal blood vessel reference area is taken as the The distance extension feature of the pixel from the normal blood vessel reference area; The absolute difference between the grayscale value of each pixel in the preset neighborhood range and the grayscale mean value of all pixels in the preset neighborhood range is recorded as the initial grayscale concentration degree. The inverse proportional value of the accumulated value of the initial grayscale concentration of all pixels within the preset neighborhood of the pixel is used as the The concentrated expression characteristics of the grayscale value within the preset neighborhood range of pixels; The grayscale mean of all pixels within the preset neighborhood of the pixel is taken as the The brightness performance characteristics within the preset neighborhood range of pixels; the distance extension characteristics, the concentration performance characteristics, and the brightness performance characteristics are fused and normalized to obtain the first pixel in the vascular ultrasound angiography image. The extension degree of each pixel from the normal blood vessel reference area; the distance extension feature, the concentration performance feature, and the brightness performance feature are all proportional to the extension degree.

[0031] As a specific example, the specific method for obtaining the extension degree is as follows: In the formula, Ultrasound angiography is the first The Euclidean distance of each pixel from the center of the normal blood vessel reference area; Ultrasound angiography is the first The number of pixels within a preset neighborhood of pixels, the preset neighborhood range in this embodiment is set to area; Ultrasound angiography is the first The first pixel in the preset neighborhood The gray value of pixels; Ultrasound angiography is the first The grayscale mean of all pixels within the preset neighborhood of pixels; is an exponential function with a natural constant as the lowest value. The model shows the inverse proportional relationship. is the input of the model; is a linear normalization function used for normalization processing; Ultrasound angiography is the first The extension degree of pixels from the normal blood vessel reference area.

[0032] It should be noted that Indicates the first The distance of pixels from normal blood vessels extends; Indicates the first The fluctuation degree of the gray value of pixels within the preset neighborhood of pixels; Indicates the first The grayscale value of the pixel in the preset neighborhood of the pixel is compared with the brightness of the background area; when the distance is extended, the fluctuation is smaller, and the brightness is greater than the background area, it means that the grayscale value of the pixel in the preset neighborhood of the pixel is compared with the brightness of the background area. The greater the extension effect of the pixel distance from the normal blood vessel reference area, the The grayscale values ​​of pixels in the preset neighborhood of pixels are relatively concentrated and are located in the blood vessel area. The blood vessel area will show a high-brightness characteristic. The greater the extension of the pixel from the normal blood vessel reference area.

[0033] Thus, the extension degree of each pixel in the vascular ultrasound contrast image from the normal blood vessel reference area is obtained.

[0034] Step S003, obtaining high-brightness pixels in a preset local area of ​​each pixel in the vascular ultrasound contrast imaging; obtaining the degree of diffusion of the preset local area where each pixel in the vascular ultrasound contrast imaging is located according to the proportion of high-brightness pixels in the preset local area, the discrete distribution between the high-brightness pixels and the degree of confusion of the pixel grayscale in the preset local area.

[0035] It should be noted that the above analysis is about the extension degree of each pixel from the normal blood vessel reference area in the vascular ultrasound contrast image, and the next step is to analyze the diffusion degree of the local area where the pixel is located in the ultrasound contrast image. Since the blood vessel is a highlighted pixel, the high-brightness pixels in the preset local area of ​​the pixel are first obtained, and then the diffusion degree of the preset local area where each pixel in the vascular ultrasound contrast image is located is obtained by analyzing the proportion of high-brightness pixels in the preset local area, the discrete distribution of high-brightness pixels, and the degree of confusion of pixel grayscale in the preset local area.

[0036] Specifically, high-brightness pixels in a preset local area of ​​each pixel in the vascular ultrasound contrast imaging are obtained as follows: For vascular ultrasound angiography The pixels within the preset local area of ​​pixels, the preset local area in this embodiment is set to In the area of ​​​​the first The preset high brightness threshold is set to 220 in this embodiment.

[0037] Preferably, in one embodiment of the present invention, the diffusion degree of the preset local area where each pixel in the vascular ultrasound contrast imaging is located is obtained according to the proportion of high-brightness pixels in the preset local area, the discrete distribution of high-brightness pixels and the degree of confusion of pixel grayscale in the preset local area. The specific method is as follows: Vascular ultrasound imaging The number of high brightness pixels in the preset local area of ​​pixels is The ratio of the number of all pixels in the preset local area of ​​pixels is taken as the first The ratio characteristics of high brightness pixels in the preset local area of ​​pixels; The variance of the Euclidean distance between high-brightness pixels in the preset local area of ​​pixels is used as the The discrete distribution characteristics between high-brightness pixels in the preset local area of ​​pixels; The information entropy of different grayscale values ​​in the preset local area of ​​pixels is used as the The chaos degree feature of the pixel grayscale in the preset local area of ​​pixels is obtained; the proportion feature, the discrete distribution feature, and the chaos degree feature are fused and normalized to obtain the first The diffusion degree of the preset local area where the pixels are located; the proportion feature is inversely proportional to the diffusion degree, and the discrete distribution feature and the chaos degree feature are directly proportional to the diffusion degree.

[0038] As a specific example, the specific method for obtaining the diffusion degree is as follows: In the formula, Ultrasound angiography is the first The number of high brightness pixels in a preset local area of ​​pixels; Ultrasound angiography is the first The number of all pixels in a preset local area of ​​pixels; Ultrasound angiography is the first The first The high brightness pixel and The Euclidean distance between high brightness pixels; Ultrasound angiography is the first The average value of the Euclidean distances between all high-brightness pixels in a preset local area of ​​pixels; Ultrasound angiography is the first The information entropy of different gray values ​​in a preset local area of ​​pixels; is a linear normalization function used for normalization processing; Ultrasound angiography is the first The degree of diffusion of the preset local area where the pixel is located.

[0039] It should be noted that Indicates the first The distribution ratio of high brightness points in the preset local area of ​​pixels, that is, the distribution of blood vessels with fine structures; Indicates the first The discrete distribution degree between high brightness pixels in the preset local area of ​​pixels. If the distribution ratio The smaller the distance distribution between high-brightness pixels, the more discrete it is, indicating that the first The higher the diffusion of pixels in the local area; Indicates the first The degree of disorder of grayscale distribution in the local area of ​​the whole pixel, The larger the value, the more chaotic the grayscale distribution is and the greater the degree of dispersion is.

[0040] Thus, the diffusion degree of the preset local area where each pixel in the vascular ultrasound contrast imaging is located is obtained.

[0041] Step S004: based on the extension degree and the diffusion degree, obtain the possibility that each pixel in the vascular ultrasound contrast image belongs to a vascular region with a fine structure; based on the possibility that the pixel belongs to a vascular region with a fine structure, obtain several vascular regions with a fine structure in the vascular ultrasound contrast image.

[0042] It should be noted that the above respectively obtains the extension degree of each pixel in the vascular ultrasound contrast image from the normal blood vessel reference area and the diffusion degree of the preset local area where each pixel in the vascular ultrasound contrast image is located, and by combining the two, several finely structured vascular areas in the vascular ultrasound contrast image are obtained.

[0043] Specifically, the possibility of each pixel in the vascular ultrasound contrast image belonging to a vascular region with a fine structure is obtained according to the extension degree and the diffusion degree, as follows: The normalized value of the product of the extension degree and the diffusion degree is used as the first The probability that a pixel belongs to a vascular region with subtle structure.

[0044] It should be noted that, in this embodiment, when normalizing the product of the extension degree and the diffusion degree, a sigmoid function is used for normalization.

[0045] Furthermore, according to the possibility that the pixel belongs to a vascular region with a fine structure, several vascular regions with a fine structure in the vascular ultrasound contrast image are obtained, as follows: Pixels whose probability of belonging to a fine-structured vascular region is greater than a preset probability threshold, which is set to 0.85 in this embodiment, are taken as pixels to be analyzed, and DBSCAN density clustering is performed on all pixels to be analyzed in the vascular ultrasound angiography image to obtain several clusters; a convex hull detection algorithm is used to perform convex hull detection on each cluster, and the minimum convex hull area corresponding to each cluster is taken as the fine-structured vascular region in the vascular ultrasound angiography image.

[0046] At this point, several vascular regions with fine structures are obtained in the vascular ultrasound contrast imaging.

[0047] Step S005, obtaining the blood vessel width area corresponding to each pixel point on the skeleton line of the fine-structured blood vessel area; according to the grayscale expression of the blood vessel width area corresponding to the pixel point on the skeleton line, the distance of the blood vessel width area from the two ends of the fine-structured blood vessel area and the possibility that the pixel belongs to the fine-structured blood vessel area, the final degree to which the blood vessel width area corresponding to each pixel in the skeleton line of each fine-structured blood vessel area belongs to the lesion is obtained.

[0048] It should be noted that before starting the enhancement analysis, since there may still be lesion areas in the fine-structured vascular area, such as thrombotic diseases, before the fine-structured vascular area is enhanced, it is still necessary to analyze the fine structure to determine whether there are lesions in the vascular tissue in the fine structure, further improve the quality of the contrast image, and achieve intelligent enhancement. After obtaining the fine-structured vascular area, if there are lesions or lesions in the vascular area, since the tissue in the lesion area usually has less contrast agent entering and the microbubble distribution in the lesion area is sparse and uneven, it appears as a persistent dark area on the ultrasound contrast image, which is specifically manifested as an interruption of inconsistent brightness at the vascular branch. Therefore, by analyzing the brightness performance of pixels in the fine-structured vascular area and combining the possibility that the pixels belong to the fine-structured vascular area, two types of pixels to be enhanced in the vascular ultrasound contrast image are obtained.

[0049] Specifically, the blood vessel width area corresponding to each pixel point on the skeleton line of the blood vessel area with fine structure is obtained as follows: Any fine-structured vascular region in the vascular ultrasound angiography image is recorded as the current fine-structured vascular region; the vascular width region corresponding to each pixel point on the skeleton line of the current fine-structured vascular region is obtained, and the vascular width region is specifically a line segment region formed by the intersection of the pixel points on the skeleton line and the edges on both sides of the current fine-structured vascular region.

[0050] Specifically, according to the grayscale expression of the blood vessel width area corresponding to the pixel point on the skeleton line, the distance between the blood vessel width area and the two ends of the blood vessel area with fine structure, and the possibility that the pixel belongs to the blood vessel area with fine structure, the final degree of the blood vessel width area corresponding to each pixel in the skeleton line of each blood vessel area with fine structure belonging to the lesion is obtained, including the steps of: First, based on the grayscale expression of the blood vessel width area corresponding to the pixel point on the skeleton line and the distance between the blood vessel width area and the two ends of the blood vessel area with fine structure, the initial degree of the blood vessel width area corresponding to each pixel in the skeleton line of each blood vessel area with fine structure belonging to the lesion is obtained, as follows: In the formula, The skeleton line of the current fine-structured blood vessel region The average gray value of the blood vessel width area corresponding to pixels; The skeleton line of the current fine-structured blood vessel region The number of pixels within a preset adjacent range of pixels, wherein the preset adjacent range is based on the The area range determined on the skeleton line with a pixel as the center and a radius of 2; is the lesion parameter, when hour, ,when hour, , is the preset first value, is the preset second value, , this implementation is based on , to give a narrative; The skeleton line of the current fine-structured blood vessel region The preset neighboring range of pixels The average gray value of the blood vessel width area corresponding to pixels; It is the average of the average grayscale values ​​of the blood vessel width area corresponding to all pixels in the skeleton line of the current fine-structured blood vessel area; The first The distance between a pixel and one end of the current fine-structured blood vessel region, where the distance can be obtained by counting the number of pixels on the skeleton line; The first The distance of pixels from the other end of the current fine-structured blood vessel area; To take the absolute value; The skeleton line of the current fine-structured blood vessel region The blood vessel width area corresponding to pixels belongs to the initial extent of the lesion.

[0051] It should be noted that the skeleton line of the current fine-structured blood vessel region The average gray value of the blood vessel width area corresponding to the pixel The smaller the The more pixels the blood vessel width area corresponds to, the more likely it is to be a lesion, because the lesion appears as a persistent dark area. While the lesion appears as a persistent dark area, it also has a certain continuity, thus presenting an area. Therefore, the average grayscale of the blood vessel width area corresponding to other pixels in its adjacent range and the overall mean of the current fine-structured blood vessel area are analyzed. For comparison, if , indicating that the average grayscale of the blood vessel width area corresponding to other pixels in its adjacent range is also low. The more pixels the blood vessel width area corresponds to, the more likely it is a lesion. Therefore, the lesion parameters are used to enlarge the area, and vice versa, the smaller the area is. Considering that when a lesion appears, the vascular area generally does not extend later, the interruption position should be close to one end of the current vascular area with fine structure. To quantify the reflection, The larger the The blood vessel width area corresponding to pixels is close to one end of the current blood vessel area with fine structure. The more likely the blood vessel width area corresponding to the pixels is to be a lesion, the greater the possibility that it belongs to the lesion.

[0052] It should be noted that the blood vessel width area corresponding to each pixel in the skeleton line of the finely structured blood vessel area obtained above belongs to the initial degree of the lesion. In order to better determine the two types of pixels to be enhanced, the possibility that the pixels in the vascular ultrasound angiography image belong to the finely structured blood vessel area is further analyzed.

[0053] Secondly, based on the possibility that the pixel belongs to the fine-structured vascular region and the initial degree to which the vascular width region corresponding to the pixel belongs to the lesion, the final degree to which the vascular width region corresponding to each pixel in the skeleton line of each fine-structured vascular region belongs to the lesion is obtained, as follows: In the formula, The skeleton line of the current fine-structured blood vessel region The blood vessel width area corresponding to pixels belongs to the initial extent of the lesion; The skeleton line of the current fine-structured blood vessel region The average probability that all pixels in the blood vessel width area corresponding to pixels belong to the blood vessel area with fine structure; To prevent the hyperparameter from having a denominator of 0, this embodiment uses to give a narrative; is a linear normalization function used for normalization processing; The skeleton line of the current fine-structured blood vessel region The blood vessel width area corresponding to pixels belongs to the final extent of the lesion.

[0054] It should be noted that if The average probability of all pixels in the blood vessel width area corresponding to pixels belonging to the fine-structured blood vessel area is smaller, indicating that it is more likely to belong to the lesion. The larger the size, the smaller the skeleton line of the current vascular region with fine structure. The more pixels the corresponding blood vessel width area is, the more likely it is to belong to the lesion, that is, Should be bigger.

[0055] At this point, the final extent to which the blood vessel width area corresponding to each pixel in the skeleton line of each finely structured blood vessel area belongs to the lesion is obtained.

[0056] Step S006, according to the final degree to which the blood vessel width area corresponding to each pixel in the skeleton line of the finely structured blood vessel area belongs to the lesion, two types of pixels to be enhanced in the vascular ultrasound angiography image are obtained; different methods are used to enhance the two types of pixels to be enhanced in the vascular ultrasound angiography image, so as to obtain an enhanced image of the patient's vascular ultrasound angiography image under a low-dose contrast agent.

[0057] It should be noted that the blood vessel width area corresponding to each pixel in the skeleton line of the finely structured blood vessel area obtained above belongs to the final extent of the lesion. By setting a suitable lesion threshold, two types of pixels to be enhanced in the vascular ultrasound angiography image are obtained.

[0058] Specifically, according to the final degree to which the blood vessel width area corresponding to each pixel in the skeleton line of the finely structured blood vessel area belongs to the lesion, two types of pixels to be enhanced in the blood vessel ultrasound contrast imaging are obtained, as follows: For all the fine-structured vascular regions in the vascular ultrasound angiography images, the pixels in the vascular width region whose final degree is greater than the preset lesion threshold are regarded as lesion-type pixels; otherwise, the pixels are regarded as fine-structured vascular pixels. The preset lesion threshold is described as 0.8 in this embodiment.

[0059] It should be noted that the above two types of pixels in the fine-structured vascular area have been obtained, and then two types of pixels to be enhanced in the entire vascular ultrasound contrast image can be obtained, namely, lesion pixels and fine-structured vascular pixels. Different enhancement strategies are used to enhance the two types of pixels to be enhanced in the vascular ultrasound contrast image, with a greater degree of enhancement for the blood vessels and a higher contrast for the lesions and background tissues, thus completing the intelligent enhancement of the vascular ultrasound contrast image.

[0060] Specifically, two types of pixels to be enhanced in the vascular ultrasound contrast image are enhanced using different methods to obtain an enhanced image of the patient's vascular ultrasound contrast image under a low-dose contrast agent, as follows: The grayscale mean of the background and the grayscale mean of the normal blood vessel reference area in the vascular ultrasound contrast-enhanced image are obtained, wherein the grayscale mean of the background is preset to 10, the grayscale value of the fine-structured blood vessel pixels in the vascular ultrasound contrast-enhanced image is set to the pixel mean of the normal blood vessel reference area, and the grayscale value of the lesion pixels in the vascular ultrasound contrast-enhanced image is set to the absolute difference between the grayscale mean of the background and the grayscale mean of the normal blood vessel reference area. times, is the preset third value. The present invention describes the enhancement of the patient's vascular ultrasound contrast imaging under low-dose contrast agent.

[0061] Through the above steps, a low-dose clinical medical ultrasound contrast intelligent enhancement method is completed.

[0062] 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. A low-dose clinical medical ultrasound contrast intelligent enhancement method, characterized in that: The method comprises the following steps: Obtain vascular ultrasound contrast images of patients under low-dose contrast agents; According to the grayscale expression of the blood vessel area in the vascular ultrasound contrast image, a normal blood vessel reference area is obtained; according to the distance extension of the pixel from the normal blood vessel reference area in the vascular ultrasound contrast image, the concentrated expression of the grayscale value within the preset neighborhood range of the pixel and the brightness expression within the preset neighborhood range, the extension degree of each pixel from the normal blood vessel reference area in the vascular ultrasound contrast image is obtained; Obtain high-brightness pixels in a preset local area of ​​each pixel in the vascular ultrasound contrast imaging; obtain the degree of diffusion of the preset local area where each pixel in the vascular ultrasound contrast imaging is located according to the proportion of high-brightness pixels in the preset local area, the discrete distribution of high-brightness pixels and the degree of confusion of pixel grayscale in the preset local area; According to the extension degree and the diffusion degree, the possibility of each pixel in the vascular ultrasound contrast image belonging to the vascular region with fine structure is obtained; according to the possibility that the pixel belongs to the vascular region with fine structure, several vascular regions with fine structure are obtained in the vascular ultrasound contrast image; Obtain the blood vessel width area corresponding to each pixel point on the skeleton line of the finely structured blood vessel area; obtain the final degree to which the blood vessel width area corresponding to each pixel point on the skeleton line of each finely structured blood vessel area belongs to the lesion according to the grayscale expression of the blood vessel width area corresponding to the pixel point on the skeleton line, the distance of the blood vessel width area from both ends of the finely structured blood vessel area, and the possibility that the pixel belongs to the finely structured blood vessel area; According to the final degree of the lesion belonging to the vascular width area corresponding to each pixel in the skeleton line of the vascular area with fine structure, two types of pixels to be enhanced in the vascular ultrasound contrast image are obtained; different methods are used to enhance the two types of pixels to be enhanced in the vascular ultrasound contrast image, and an enhanced image of the vascular ultrasound contrast image of the patient under low-dose contrast agent is obtained; The method of obtaining the extension degree of each pixel in the vascular ultrasound contrast image from the normal blood vessel reference area according to the distance extension of the pixel in the vascular ultrasound contrast image from the normal blood vessel reference area, the concentrated expression of the gray value in the preset neighborhood range of the pixel, and the brightness expression in the preset neighborhood range includes the following specific steps: Vascular ultrasound imaging The Euclidean distance of the pixel from the center of the normal blood vessel reference area is taken as the The distance extension feature of the pixel from the normal blood vessel reference area; The absolute difference between the grayscale value of each pixel in the preset neighborhood range and the grayscale mean value of all pixels in the preset neighborhood range is recorded as the initial grayscale concentration degree. The inverse proportional value of the accumulated value of the initial grayscale concentration of all pixels within the preset neighborhood of the pixel is used as the The concentrated expression characteristics of the grayscale value within the preset neighborhood range of pixels; The grayscale mean of all pixels within the preset neighborhood of the pixel is taken as the The brightness performance characteristics within the preset neighborhood range of pixels; the distance extension characteristics, the concentration performance characteristics, and the brightness performance characteristics are fused and normalized to obtain the first pixel in the vascular ultrasound angiography image. The extension degree of each pixel from the normal blood vessel reference area; the distance extension feature, the concentration performance feature, and the brightness performance feature are all proportional to the extension degree; The specific steps of obtaining the diffusion degree of the preset local area where each pixel in the vascular ultrasound contrast imaging is located according to the proportion of high-brightness pixels in the preset local area, the discrete distribution of high-brightness pixels and the degree of confusion of pixel grayscale in the preset local area are as follows: Vascular ultrasound imaging The number of high brightness pixels in the preset local area of ​​pixels is The ratio of the number of all pixels in the preset local area of ​​pixels is taken as the first The ratio characteristics of high brightness pixels in the preset local area of ​​pixels; The variance of the Euclidean distance between high-brightness pixels in the preset local area of ​​pixels is used as the The discrete distribution characteristics between high-brightness pixels in the preset local area of ​​pixels; The information entropy of different grayscale values ​​in the preset local area of ​​pixels is used as the The chaos degree feature of the pixel grayscale in the preset local area of ​​pixels is obtained; the proportion feature, the discrete distribution feature, and the chaos degree feature are fused and normalized to obtain the first The diffusion degree of the preset local area where the pixels are located; the proportion feature is inversely proportional to the diffusion degree, and the discrete distribution feature and the chaos degree feature are directly proportional to the diffusion degree; The specific method for obtaining the blood vessel width area corresponding to each pixel point on the skeleton line of the blood vessel area with fine structure is as follows: Record any fine-structured vascular region in the vascular ultrasound contrast imaging as the current fine-structured vascular region; obtain the vascular width region corresponding to each pixel point on the skeleton line of the current fine-structured vascular region, wherein the vascular width region is a line segment region formed by the intersection of the pixel point on the skeleton line and the edges on both sides of the current fine-structured vascular region; The method of obtaining the final degree of the blood vessel width area corresponding to each pixel in the skeleton line of each fine-structured blood vessel area belonging to the lesion according to the grayscale expression of the blood vessel width area corresponding to the pixel point on the skeleton line, the distance between the blood vessel width area and the two ends of the fine-structured blood vessel area, and the possibility that the pixel belongs to the fine-structured blood vessel area, includes the following specific steps: In the formula, The skeleton line of the current fine-structured blood vessel region The average gray value of the blood vessel width area corresponding to pixels; The skeleton line of the current fine-structured blood vessel region The number of pixels within a preset neighborhood of pixels; is the lesion parameter, when hour, ,when hour, , is the preset first value, is the preset second value, ; The skeleton line of the current fine-structured blood vessel region The preset neighboring range of pixels The average gray value of the blood vessel width area corresponding to pixels; It is the average of the average grayscale values ​​of the blood vessel width area corresponding to all pixels in the skeleton line of the current fine-structured blood vessel area; The first The distance of pixels from one end of the current fine-structured blood vessel area; The first The distance of pixels from the other end of the current fine-structured blood vessel area; To take the absolute value; The skeleton line of the current fine-structured blood vessel region The blood vessel width area corresponding to pixels belongs to the initial extent of the lesion; According to the possibility that the pixel belongs to the fine-structured vascular region and the initial degree to which the vascular width region corresponding to the pixel belongs to the lesion, the final degree to which the vascular width region corresponding to each pixel in the skeleton line of each fine-structured vascular region belongs to the lesion is obtained; The method of obtaining the final degree to which the blood vessel width area corresponding to each pixel in the skeleton line of each blood vessel area with fine structure belongs to the lesion according to the possibility that the pixel belongs to the blood vessel area with fine structure and the initial degree to which the blood vessel width area corresponding to the pixel belongs to the lesion comprises the following specific steps: In the formula, The skeleton line of the current fine-structured blood vessel region The average probability that all pixels in the blood vessel width area corresponding to pixels belong to the blood vessel area with fine structure; To prevent hyperparameters with denominators equal to 0; is a linear normalization function; The skeleton line of the current fine-structured blood vessel region The blood vessel width area corresponding to pixels belongs to the final extent of the lesion.

2. According to claim 1, a low-dose clinical medical ultrasound contrast intelligent enhancement method is characterized in that: The method of obtaining the possibility that each pixel in the vascular ultrasound contrast imaging belongs to a vascular region with a fine structure according to the extension degree and the diffusion degree includes the following specific steps: The normalized value of the product of the extension degree and the diffusion degree is used as the first The probability that a pixel belongs to a vascular region with subtle structure.

3. According to claim 1, a low-dose clinical medical ultrasound contrast intelligent enhancement method is characterized in that: The method of obtaining a plurality of vascular regions with fine structures in the vascular ultrasound contrast imaging according to the possibility that the pixel belongs to the vascular region with fine structures comprises the following specific steps: Pixels whose probability of belonging to the subtle vascular area is greater than the preset probability threshold are taken as pixels to be analyzed, and DBSCAN density clustering is performed on all pixels to be analyzed in the vascular ultrasound contrast images to obtain several clusters. The convex hull detection algorithm is used to perform convex hull detection on each cluster, and the minimum convex hull area corresponding to each cluster is taken as the subtle vascular area in the vascular ultrasound contrast images.

4. According to claim 1, a low-dose clinical medical ultrasound contrast intelligent enhancement method is characterized in that: The method of obtaining two types of pixels to be enhanced in the vascular ultrasound contrast imaging according to the final extent to which the vascular width area corresponding to each pixel in the skeleton line of the finely structured vascular area belongs to the lesion includes the following specific steps: For all the fine-structured vascular regions in the vascular ultrasound contrast imaging, the pixels in the vascular width region whose final degree is greater than the preset lesion threshold are regarded as lesion-type pixels; otherwise, they are regarded as fine-structured vascular-type pixels.

5. According to claim 1, a low-dose clinical medical ultrasound contrast intelligent enhancement method is characterized in that: The method of enhancing the two types of pixels to be enhanced in the vascular ultrasound contrast-enhanced image by different methods to obtain an enhanced image of the vascular ultrasound contrast-enhanced image of the patient under low-dose contrast agent includes the following specific steps: The grayscale mean of the background and the grayscale mean of the normal blood vessel reference area in the vascular ultrasound contrast-enhanced image are obtained, wherein the grayscale mean of the background is a preset parameter, and the grayscale value of the blood vessel-like pixels with fine structure in the vascular ultrasound contrast-enhanced image is set to the pixel mean of the normal blood vessel reference area, and the grayscale value of the lesion-like pixels in the vascular ultrasound contrast-enhanced image is set to the absolute difference between the grayscale mean of the background and the grayscale mean of the normal blood vessel reference area. times, is a preset third value, and an enhanced image of the patient's vascular ultrasound contrast imaging under a low-dose contrast agent is obtained.

Images