A method for automatically acquiring cardiac ultrasound images

Abstract

The present invention relates to the technical field of image enhancement, and in particular to an automatic acquisition method for cardiac ultrasound images, which screens out suspected calcified pixel points according to the characteristics of highlighted pixel points of tissue calcification, and screens out suspected artifact points that may be affected based on the imaging principle; further determines the artifact feature area of ​​each suspected artifact point based on the grayscale similarity characteristics based on the artifact area, and determines the artifact possibility according to the characteristics that the grayscale attenuation and the law of the unchanged artifact area will be destroyed after the artifact feature area is affected by other factors; determines the grayscale correction weight that characterizes the degree of influence of the artifact based on the grayscale gradient trend and the characteristics that the morphology of the artifact area remains unchanged, and corrects the grayscale value of each suspected artifact point according to the grayscale correction weight, thereby reducing the influence of the artifact on the cardiac ultrasound image, making the acquired enhanced cardiac ultrasound enhanced image more accurate, thereby improving the effect of medical teaching based on the cardiac ultrasound enhanced image.

Description

Technical Field

[0001] The present invention relates to the technical field of image enhancement, and in particular to a method for automatically acquiring cardiac ultrasonic images. Background Art

[0002] Cardiac ultrasound is a technique for observing the structure and function of the heart in detail through ultrasonic imaging. During the acquisition of cardiac ultrasound images, when ultrasound waves are emitted from the probe, the reflected waves are directly transmitted back to the probe, making a round trip, thus forming an image on the cardiac ultrasound image. As the structure and function of the heart can be observed in detail, cardiac ultrasound images are often used as teaching materials in medical teaching. In order to improve the effect of medical teaching, image enhancement operations are usually required after acquiring cardiac ultrasound images.

[0003] The prior art usually directly performs histogram equalization on the acquired cardiac ultrasound image through histogram equalization to obtain an enhanced cardiac ultrasound enhanced image, and performs better medical teaching based on the cardiac ultrasound enhanced image; however, in the actual process of acquiring the cardiac ultrasound image, due to the influence of certain tissue calcification, the reflected wave may form multiple reflections between the calcified tissue and the tissue to be imaged, so that the ultrasound probe receives the image signal of the tissue to be imaged multiple times, resulting in the formation of a step-like artifact in the obtained cardiac ultrasound image. The influence of the step-like artifact cannot be effectively removed by the histogram equalization method alone, resulting in poor effect of image enhancement through histogram equalization, and poor effect of medical teaching directly based on the acquired cardiac ultrasound enhanced image. Summary of the invention

[0004] The present application provides a method for automatically acquiring cardiac ultrasound images. The method first selects suspected calcified pixels based on the characteristics of highlighted pixels of tissue calcification, and selects suspected artifact points that may be affected based on the imaging principle; further determines the artifact feature area of ​​each suspected artifact point based on the grayscale similarity characteristics based on the artifact area, and determines the possibility of artifacts based on the characteristics that the grayscale attenuation and the law of unchanged artifact area will be destroyed after the artifact feature area is affected by other factors; then determines the grayscale correction weight that characterizes the degree of influence of the artifact based on the grayscale gradient trend and the characteristics of unchanged morphology of the artifact area, and corrects the grayscale value of each suspected artifact point according to the grayscale correction weight, thereby reducing the influence of the artifact on the cardiac ultrasound image, solving the problem that the influence of step-type artifacts cannot be effectively removed only by the histogram equalization method, resulting in poor effect of image enhancement by histogram equalization, so that the acquired enhanced cardiac ultrasound enhanced image is more accurate, thereby improving the effect of medical teaching based on the cardiac ultrasound enhanced image.

[0005] The present application provides a method for automatically acquiring cardiac ultrasound images, comprising:

[0006] Acquire a cardiac ultrasound grayscale image of the patient through an ultrasound probe; perform threshold segmentation based on the grayscale value distribution in the cardiac ultrasound grayscale image to screen out suspected calcified pixels; determine a corresponding suspected artifact point sequence based on the position of the suspected calcified pixels relative to the corresponding pixel of the ultrasound probe;

[0007] Determine the reference degree of the artifact region between each pixel point and each suspected artifact point according to the similarity of the grayscale values ​​of all pixels between each pixel point and each suspected artifact point; determine the corresponding artifact feature region according to the reference degree of the artifact region of each pixel point corresponding to each suspected artifact point;

[0008] In the sequence of suspected artifact points, the anti-interference degree of each suspected artifact point is determined according to the compliance of the grayscale attenuation law between each suspected artifact point and an adjacent suspected artifact point and the similarity of the area of ​​the artifact feature region; the artifact credibility is determined based on the minimum value of the anti-interference degree between each suspected artifact point and an adjacent suspected artifact point;

[0009] According to the overall grayscale attenuation trend of the artifact feature area between adjacent suspected artifact points combined with the corresponding artifact credibility, the artifact grayscale gradient of all suspected artifact points is determined; according to the morphological similarity of the artifact feature area between adjacent suspected artifact points combined with the corresponding artifact credibility, the artifact morphological similarity of each suspected artifact point is determined;

[0010] According to the grayscale gradient of the artifact and the similarity of the artifact morphology, a grayscale correction weight of each suspected artifact point is determined; the grayscale value of each suspected artifact point is corrected according to the grayscale correction weight to determine an enhanced cardiac ultrasound enhanced image.

[0011] Furthermore, the process of acquiring the suspected calcified pixel points includes:

[0012] The grayscale threshold is determined by the Otsu threshold method according to the grayscale values ​​of all pixels in the cardiac ultrasound grayscale image; and the pixels whose grayscale values ​​are greater than the grayscale threshold are regarded as suspected calcification pixels.

[0013] Furthermore, the process of acquiring the suspected false image point sequence includes:

[0014] The pixel point corresponding to the position of the ultrasound probe is used as the ultrasound probe point; and the artifact interval distance of the suspected calcification pixel point is determined according to the Euclidean distance between the ultrasound probe point and the suspected calcification pixel point;

[0015] On a ray pointing to the suspected calcified pixel point with the ultrasound probe point as the endpoint, a suspected artifact point is selected at every artifact interval distance; all suspected artifact points are arranged in sequence along the extension direction of the ray to determine a suspected artifact point sequence corresponding to the suspected calcified pixel point.

[0016] Furthermore, the process of obtaining the reference degree of the artifact area includes:

[0017] Taking each suspected artifact point as a target artifact point in turn; calculating the grayscale value variance of all pixels on the line between each pixel point in the cardiac ultrasound grayscale image and the target artifact point; performing negative correlation mapping on the difference between the grayscale value of each pixel point in the cardiac ultrasound grayscale image and the grayscale value of the target artifact point, and determining the reference grayscale similarity of each pixel point;

[0018] The product of the negative correlation mapping value of the gray value variance and the reference gray similarity is normalized to determine the artifact region reference degree between each pixel point and the target artifact point in the cardiac ultrasound gray image.

[0019] Furthermore, the process of acquiring the artifact feature area includes:

[0020] Determine a reference degree threshold value by using the Otsu threshold method according to the reference degrees of all artifact regions corresponding to each suspected artifact point; use a pixel point whose artifact region reference degree between each suspected artifact point and the pixel point is greater than the reference degree threshold value as an artifact feature point of each suspected artifact point;

[0021] The connected domain composed of all the artifact feature points corresponding to each suspected artifact point in the cardiac ultrasound grayscale image is used as the reference feature region of each suspected artifact point; and the reference feature region where each suspected artifact point is located is used as the corresponding artifact feature region.

[0022] Furthermore, the process of obtaining the anti-interference degree includes:

[0023] In the suspected artifact point sequence, negative correlation mapping is performed on the difference between the grayscale value of each suspected artifact point and the grayscale value of the previous suspected artifact point to determine the conformity degree of the attenuation law of each suspected artifact point;

[0024] Determine the area difference of the feature region of each suspected artifact point according to the difference between the area of ​​the artifact feature region of each suspected artifact point and the area of ​​the artifact feature region of the last suspected artifact point;

[0025] The product of the negative correlation mapping value of the area difference of the characteristic region and the degree of compliance with the attenuation law is normalized to determine the anti-interference degree of each suspected artifact point.

[0026] Furthermore, the process of obtaining the grayscale gradient of the artifact image includes:

[0027] The mean grayscale value of all pixels in the artifact feature area of ​​each suspected artifact point is taken as the artifact grayscale feature value of each suspected artifact point;

[0028] In the sequence of suspected artifact points, the negative correlation mapping value of the difference between the artifact grayscale eigenvalue of each suspected artifact point and the artifact grayscale eigenvalue of the previous suspected artifact point is used as the grayscale gradient eigenvalue of each suspected artifact point; and the product of the artifact credibility and the grayscale gradient eigenvalue is used as the artifact grayscale gradient of each suspected artifact point.

[0029] Furthermore, the process of obtaining the artifact morphology similarity includes:

[0030] In the artifact feature area of ​​each suspected artifact point, the pixel point closest to the vertex position of the upper left corner of the cardiac ultrasound grayscale image is taken as the starting point, and the Freeman chain code of the corresponding artifact feature area is obtained by the chain code method;

[0031] In the suspected artifact point sequence, a DTW distance between a Freeman chain code corresponding to each suspected artifact point and a Freeman chain code corresponding to a previous suspected artifact point is calculated by a dynamic time warping algorithm;

[0032] The product of the DTW distance and the area difference of the feature region is negatively correlated to determine the feature region similarity of each suspected artifact point; the product of the feature region similarity and the artifact credibility is used as the artifact morphology similarity of each suspected artifact point.

[0033] Furthermore, the grayscale correction weight acquisition process includes:

[0034] The product of the grayscale gradient of the artifact and the morphological similarity of the artifact is normalized to determine the grayscale correction weight of each suspected artifact point.

[0035] Furthermore, the process of acquiring the cardiac ultrasound enhanced image includes:

[0036] Count all suspected artifact points corresponding to all suspected calcified pixel points; in the cardiac ultrasound grayscale image, take the product of the negative correlation mapping value of the grayscale correction weight of each suspected artifact point and the corresponding grayscale value as the enhanced grayscale value of each suspected artifact point; replace the grayscale values ​​of all suspected artifact points in the cardiac ultrasound grayscale image with the corresponding enhanced grayscale values, and determine the enhanced cardiac ultrasound enhanced image.

[0037] This application has the following beneficial effects:

[0038] The step-like artifacts in cardiac ultrasound images are caused by multiple reflections between the calcified tissue and the tissue to be imaged, so the artifact pixels exist on the extension line from the ultrasound probe to the calcified pixel, and due to the influence of reflection, the artifacts appear regularly at equal distances; therefore, after screening out the suspected calcified pixels with higher grayscale values, the sequence of suspected artifact points generated under the influence of each suspected calcified pixel is determined based on the position of the suspected calcified pixel relative to the corresponding pixel of the ultrasound probe.

[0039] The artifact area is produced by multiple reflections of the same tissue to be imaged, so each artifact area represents the tissue to be imaged, so the grayscale value in each artifact area usually shows a high degree of consistency; therefore, according to the similarity of the grayscale values ​​of all pixels between each pixel point and each suspected artifact point, the obtained artifact area reference degree characterizes the possibility that the corresponding pixel point belongs to the artifact feature area pixel point, thereby determining the artifact feature area where each suspected artifact point is located based on the artifact area reference degree of each pixel point.

[0040] Taking into account that after the artifact area is superimposed with the ultrasonic image of the real tissue, the corresponding artifact area will adhere to the real ultrasonic image area, causing the corresponding artifact area to be affected; in order to determine the imaging effect of the artifact area to better correct the grayscale value, it is first necessary to determine the interference or resistance to interference of each artifact area; for the artifacts of step imaging, as the reflection proceeds, the grayscale value of each artifact area will gradually decay, but the shape usually does not change. If this law is destroyed, it means that the artifact area is interfered with and the corresponding anti-interference degree is small; therefore, the present application determines the anti-interference degree of each suspected artifact point according to the compliance of the grayscale attenuation law between each suspected artifact point and the adjacent suspected artifact point and the similarity of the area of ​​the artifact feature area, thereby determining the credibility of the generated artifact, that is, the artifact credibility, according to the anti-interference degree.

[0041] Since the step artifact is formed by multiple reflections of the reflected wave between the calcified tissue and the tissue to be imaged, there is energy loss in the reflected wave during the reflection process, resulting in the reflection intensity becoming weaker and weaker, which is manifested as the grayscale value of the artifact feature area becoming smaller and smaller in the cardiac ultrasound image. Therefore, as the number of reflections increases, the overall grayscale value of the artifact feature area gradually decreases. Therefore, for each suspected calcified pixel point, if the artifact feature area between it and the previous suspected calcified pixel point in the suspected artifact point sequence meets the characteristics of the overall grayscale attenuation trend, it means that the corresponding suspected calcified pixel point is more consistent with the artifact feature in the dimension of grayscale gradient. Therefore, on this basis, combined with the artifact credibility, the more accurate artifact grayscale gradient characterizing the artifact probability corresponding to each suspected artifact point is determined; and the artifact area is formed by the reflection of ultrasound by the tissue to be imaged, so the corresponding artifact areas should have a high degree of morphological similarity; therefore, on this basis, combined with the artifact credibility, the artifact morphological similarity characterizing the artifact probability is determined.

[0042] Finally, the grayscale correction weight of each suspected artifact point is determined based on the artifact grayscale gradient that characterizes the artifact probability and the artifact morphology similarity, so that the grayscale value of each suspected artifact point is further corrected according to the grayscale correction weight, thereby reducing the influence of the artifact on the cardiac ultrasound image and making the acquired enhanced cardiac ultrasound enhanced image more accurate, thereby improving the effect of medical teaching based on the cardiac ultrasound enhanced image. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are 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.

[0044] Figure 1 A flow chart of a method for automatically acquiring cardiac ultrasound images provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0045] 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 a method for automatic acquisition of cardiac ultrasound images proposed according to the present invention, its specific implementation method, structure, features 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, and the specific features, structures or characteristics in one or more embodiments may be combined in any suitable form. In addition, the terms "first" and "second" are used only for descriptive purposes and cannot be understood as implying or suggesting relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features.

[0046] 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.

[0047] The specific scheme of the method for automatically acquiring cardiac ultrasonic images provided by the present invention is described in detail below with reference to the accompanying drawings.

[0048] This application embodiment provides a method for automatically acquiring cardiac ultrasound images. Figure 1 , which shows a flow chart of a method for automatically acquiring cardiac ultrasound images provided by an embodiment of the present invention, the method comprising:

[0049] Step S101: Acquire a cardiac ultrasound grayscale image of the patient through an ultrasound probe; perform threshold segmentation to screen out suspected calcified pixels according to the grayscale value distribution in the cardiac ultrasound grayscale image; and determine a corresponding suspected artifact point sequence according to the position of the suspected calcified pixels relative to the corresponding pixels of the ultrasound probe.

[0050] Bring the patient into the examination room and make the patient lie on the examination bed in the left lateral position, prepare the ultrasound equipment and select the corresponding ultrasound probe, apply coupling agent on the ultrasound probe, place the probe next to the left sternum, adjust the angle of the probe, collect the cardiac ultrasound image of the parasternal long axis section, further grayscale the cardiac ultrasound image, and obtain the cardiac ultrasound grayscale image required for this application.

[0051] The staircase artifacts in cardiac ultrasound images are caused by multiple reflections between calcified tissue and the tissue to be imaged, so the artifact pixels exist on the extension line from the ultrasound probe to the calcified pixels. Therefore, before determining the artifact area, it is necessary to first obtain the suspected calcified pixels corresponding to the calcified area. Calcified tissue appears as a high-brightness area in cardiac ultrasound images, so the grayscale values ​​of suspected calcified pixels are usually relatively high. Therefore, threshold segmentation is performed based on the grayscale value distribution in the cardiac ultrasound grayscale image to screen out suspected calcified pixels.

[0052] Preferably, in some possible implementations of the embodiments of the present invention, the process of acquiring suspected calcified pixels includes:

[0053] The grayscale threshold is determined by the Otsu threshold method according to the grayscale values ​​of all pixels in the cardiac ultrasound grayscale image; pixels greater than the grayscale threshold are regarded as suspected calcified pixels. It should be noted that the Otsu threshold method is a technical means well known to those skilled in the art and will not be further defined or elaborated herein.

[0054] Ultrasonic waves are emitted by an ultrasonic probe. They are reflected by the tissue to be imaged to form reflected waves. The reflected waves are reflected multiple times between the calcified tissue and the tissue to be imaged, causing the ultrasonic probe to receive signals from the tissue to be imaged multiple times, thereby forming a step-like artifact. That is, the artifact areas will be evenly spaced on the cardiac ultrasound image, and the interval length is determined by the distance between the calcified tissue and the ultrasonic probe. Therefore, the corresponding suspected artifact point sequence is further determined based on the position of the suspected calcified pixel point relative to the corresponding pixel point of the ultrasonic probe, that is, the artifact point suspected to be affected by the corresponding suspected calcified pixel point.

[0055] Preferably, in some possible implementations of the embodiments of the present invention, the process of acquiring the suspected artifact point sequence includes:

[0056] The pixel point corresponding to the position of the ultrasound probe is used as the ultrasound probe point; the artifact interval distance of the suspected calcification pixel point is determined according to the Euclidean distance between the ultrasound probe point and the suspected calcification pixel point; on the ray with the ultrasound probe point as the endpoint and pointing to the suspected calcification pixel point, a suspected artifact point is selected every artifact interval distance; all suspected artifact points are arranged in sequence in the extension direction of the ray to determine the suspected artifact point sequence corresponding to the suspected calcification pixel point.

[0057] Step S102: Determine the degree of reference of the artifact area between each pixel and each suspected artifact point according to the similarity of the grayscale values ​​of all pixels between each pixel and each suspected artifact point; determine the corresponding artifact feature area according to the degree of reference of the artifact area of ​​each pixel corresponding to each suspected artifact point.

[0058] The artifact area is produced by multiple reflections of the same tissue to be imaged, so each artifact area represents the tissue to be imaged, so the grayscale value in each artifact area usually shows a high degree of consistency; therefore, according to the similarity of the grayscale values ​​of all pixels between each pixel point and each suspected artifact point, the obtained artifact area reference degree characterizes the possibility that the corresponding pixel point belongs to the artifact feature area pixel point, thereby determining the artifact feature area where each suspected artifact point is located based on the artifact area reference degree of each pixel point.

[0059] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the reference degree of the artifact area includes:

[0060] Each suspected artifact point is taken as a target artifact point in turn; the grayscale value variance of all pixels on the line between each pixel point and the target artifact point in the cardiac ultrasound grayscale image is calculated; the difference between the grayscale value of each pixel point in the cardiac ultrasound grayscale image and the grayscale value of the target artifact point is negatively correlated and mapped to determine the reference grayscale similarity of each pixel point. For each pixel point in the cardiac ultrasound grayscale image, the smaller the grayscale value variance of all pixels between it and the target artifact point, the more uniform the grayscale value distribution between the corresponding pixel point and the target artifact point, which means that the pixel point and the target artifact point are more likely to belong to the same artifact area with uniform grayscale value, and the corresponding artifact area reference degree should be greater; similarly, the closer the grayscale value between the corresponding pixel point and the target artifact point, that is, the greater the reference grayscale similarity, the more likely the pixel point and the target artifact point are to belong to the same artifact area with uniform grayscale value, and the corresponding artifact area reference degree should be greater.

[0061] Therefore, for each pixel point, its corresponding gray value variance is negatively correlated with the reference degree of the artifact area, while the reference gray similarity is positively correlated with the reference degree of the artifact area; further, the product between the negative correlation mapping value of the gray value variance and the reference gray similarity is normalized to determine the reference degree of the artifact area between each pixel point in the cardiac ultrasound grayscale image and the target artifact point.

[0062] In a specific implementation of the embodiment of the present invention, the process of obtaining the reference degree of the artifact area is expressed by the formula: ;in, The first Pixels and suspected artifacts The reference degree of the artifact area between; The first Pixels and suspected artifacts The gray value variance of all pixels on the line between ; Suspected artifact point Gray value of The first The gray value of each pixel; is the absolute value symbol; is the linear normalization function, The first Pixels and suspected artifacts Grayscale similarity between ; is an exponential function with a natural constant as base; is a normalization function, and the sum of all the normalized values ​​corresponding to it is 1, which is also the suspected artifact point The sum of the reference degrees of the artifact areas between all pixels in the cardiac ultrasound grayscale image is 1.

[0063] Preferably, in some possible implementations of the embodiments of the present invention, the process of acquiring the artifact feature region includes:

[0064] The reference degree threshold is determined by the Otsu threshold method according to the reference degree of all artifact areas corresponding to each suspected artifact point; the pixel points whose artifact area reference degree between each suspected artifact point is greater than the reference degree threshold are used as the artifact feature points of each suspected artifact point; the artifact feature points with a larger reference degree of the artifact area are screened and processed by the reference degree threshold, and for each real artifact point, its corresponding artifact area presents a high grayscale consistency, so the pixels in its corresponding artifact feature area all correspond to a larger artifact area reference degree, that is, when the corresponding suspected artifact point is a real artifact point, all the pixels in its corresponding artifact area are usually artifact feature points, so according to this feature, the connected domain composed of all artifact feature points corresponding to each suspected artifact point in the cardiac ultrasound grayscale image is used as the reference feature area of ​​each suspected artifact point; the reference feature area where each suspected artifact point is located is used as the corresponding artifact feature area.

[0065] Step S103: In the sequence of suspected artifact points, the anti-interference degree of each suspected artifact point is determined according to the conformity of the grayscale attenuation law between each suspected artifact point and the adjacent suspected artifact points and the similarity of the area of ​​the artifact feature region; the artifact credibility is determined based on the minimum value of the anti-interference degree between each suspected artifact point and the adjacent suspected artifact points.

[0066] Taking into account that after the artifact area is superimposed with the ultrasonic image of the real tissue, the corresponding artifact area will adhere to the real ultrasonic image area, causing the corresponding artifact area to be affected; in order to determine the imaging effect of the artifact area to better correct the grayscale value, it is first necessary to determine the interference or resistance to interference of each artifact area; for the artifacts of step imaging, as the reflection proceeds, the grayscale value of each artifact area will gradually decay, but the shape usually does not change. If this law is destroyed, it means that the artifact area is interfered with and the corresponding anti-interference degree is small; therefore, the present application determines the anti-interference degree of each suspected artifact point according to the compliance of the grayscale attenuation law between each suspected artifact point and the adjacent suspected artifact point and the similarity of the area of ​​the artifact feature area, thereby determining the credibility of the generated artifact, that is, the artifact credibility, according to the anti-interference degree.

[0067] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the anti-interference degree includes:

[0068] In the sequence of suspected artifact points, the difference between the grayscale value of each suspected artifact point and the grayscale value of the previous suspected artifact point is negatively correlated to determine the degree of compliance with the attenuation law of each suspected artifact point. With the characteristic that the grayscale value of the step-type artifact will gradually decay, if the grayscale value of each suspected artifact point is greater than the grayscale value of the previous suspected artifact point, it means that the grayscale attenuation characteristics of the suspected artifact point are destroyed, that is, the smaller the impact on the corresponding artifact area, the greater the corresponding anti-interference degree. Therefore, the larger the negative correlation mapping value of the difference between the grayscale value of each suspected artifact point and the grayscale value of the previous suspected artifact point, the greater the degree of compliance with the corresponding attenuation law, that is, the greater the anti-interference degree.

[0069] The area difference of the feature area of ​​each suspected artifact point is determined based on the difference between the area of ​​the artifact feature area of ​​each suspected artifact point and the area of ​​the artifact feature area of ​​the previous suspected artifact point. Since the area of ​​the artifact area usually does not change, the larger the difference in the feature area, the more likely it is to be affected, and the smaller the corresponding anti-interference degree.

[0070] Since the smaller the difference in the area of ​​the feature region is and the greater the degree of compliance with the attenuation law is, the greater the anti-interference degree of the corresponding suspected artifact point is, the product of the negative correlation mapping value of the difference in the area of ​​the feature region and the degree of compliance with the attenuation law is further normalized to determine the anti-interference degree of each suspected artifact point. It should be noted that for the first suspected artifact point in the suspected artifact point sequence, since there is no previous suspected artifact point, in order to enable the subsequent analysis to proceed smoothly, the average of the anti-interference degrees of other suspected artifact points other than the first suspected artifact point in the suspected artifact point sequence is used as the anti-interference degree of the first suspected artifact point.

[0071] In a specific implementation of the embodiment of the present invention, the process of obtaining the anti-interference degree is expressed by the formula: ;in, For the In the sequence of suspected artifact points of suspected calcified pixels, The anti-interference degree of the suspected artifact point; For the In the sequence of suspected artifact points of suspected calcified pixels, Gray value of suspected artifact points; For the In the sequence of suspected artifact points of suspected calcified pixels, Gray value of suspected artifact points; For the In the sequence of suspected artifact points of suspected calcified pixels, The degree of conformity of the attenuation law of the suspected artifact points; is an exponential function with a natural constant as base; For the In the sequence of suspected artifact points of suspected calcified pixels, The area of ​​the artifact feature region of the suspected artifact point; For the In the sequence of suspected artifact points of suspected calcified pixels, The area of ​​the artifact feature region of the suspected artifact point; For the In the sequence of suspected artifact points of suspected calcified pixels, The difference in the area of ​​the characteristic regions of the suspected artifact points; is the hyperbolic tangent function.

[0072] In a specific implementation method of an embodiment of the present invention, the minimum anti-interference degree between each suspected artifact point and the previous suspected artifact point is used as the artifact credibility of each suspected artifact point. Because the subsequent artifact probability analysis is based on the two suspected artifact points, in order to ensure the accuracy of the subsequent analysis, only when the anti-interference degree of the two suspected artifact points is large, can it be considered that the subsequent analysis is under the influence of artifacts. Therefore, the minimum anti-interference degree is selected as the artifact credibility.

[0073] Step S104: Determine the artifact grayscale gradient of all suspected artifact points based on the overall grayscale attenuation trend of the artifact feature area between adjacent suspected artifact points combined with the corresponding artifact credibility; determine the artifact morphological similarity of each suspected artifact point based on the morphological similarity of the artifact feature area between adjacent suspected artifact points combined with the corresponding artifact credibility.

[0074] Since the step artifact is formed by multiple reflections of the reflected wave between the calcified tissue and the tissue to be imaged, there is energy loss in the reflected wave during the reflection process, resulting in the reflection intensity becoming weaker and weaker, which is manifested as the grayscale value of the artifact feature area becoming smaller and smaller in the cardiac ultrasound image. Therefore, as the number of reflections increases, the overall grayscale value of the artifact feature area gradually decreases. Therefore, for each suspected calcified pixel point, if the artifact feature area between it and the previous suspected calcified pixel point in the suspected artifact point sequence meets the characteristics of the overall grayscale attenuation trend, it means that the corresponding suspected calcified pixel point is more consistent with the artifact feature in the dimension of grayscale gradient. Therefore, on this basis, combined with the artifact credibility, the more accurate artifact grayscale gradient characterizing the artifact probability corresponding to each suspected artifact point is determined; and the artifact area is formed by the reflection of ultrasound by the tissue to be imaged, so the corresponding artifact areas should have a high degree of morphological similarity; therefore, on this basis, combined with the artifact credibility, the artifact morphological similarity characterizing the artifact probability is determined.

[0075] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the grayscale gradient of the artifact includes:

[0076] The grayscale value average of all pixels in the artifact feature area of ​​each suspected artifact point is used as the artifact grayscale feature value of each suspected artifact point; in the suspected artifact point sequence, the negative correlation mapping value of the difference between the artifact grayscale feature value of each suspected artifact point and the artifact grayscale feature value of the previous suspected artifact point is used as the grayscale gradient feature value of each suspected artifact point. Similar to the acquisition process of the degree of compliance with the attenuation law, when the corresponding artifact area is a real artifact, due to the energy loss of the reflected wave during the reflection process, the artifact grayscale feature value of the corresponding suspected artifact point is smaller than the artifact grayscale feature value of the previous suspected artifact point; therefore, the larger the grayscale gradient feature value, the more consistent with the feature that the artifact grayscale feature value of the suspected artifact point is smaller than the artifact grayscale feature value of the previous suspected artifact point, the more the corresponding suspected artifact point conforms to the grayscale gradient feature of the artifact area, and the higher the corresponding artifact probability. Since the artifact credibility can also characterize the corresponding artifact probability, the artifact credibility is further combined with the grayscale gradient eigenvalue, and the product of the artifact credibility and the grayscale gradient eigenvalue is used as the artifact grayscale gradient of each suspected artifact point, so that the larger the artifact grayscale gradient, the larger the corresponding artifact probability.

[0077] In a specific implementation of the embodiment of the present invention, the process of obtaining the grayscale gradient of the artifact image is expressed by the formula: ;in, For the In the sequence of suspected artifact points of suspected calcified pixels, The grayscale gradient of the suspected artifact points; For the In the sequence of suspected artifact points of suspected calcified pixels, The credibility of the artifacts of the suspected artifact points; For the In the sequence of suspected artifact points of suspected calcified pixels, The mean gray value of all pixels in the artifact feature area of ​​the suspected artifact point, that is, the artifact gray feature value; For the In the sequence of suspected artifact points of suspected calcified pixels, The mean gray value of all pixels in the artifact feature area of ​​the suspected artifact point, that is, the artifact gray feature value; is an exponential function with a natural constant as base; For the In the sequence of suspected artifact points of suspected calcified pixels, The grayscale gradient feature value of a suspected artifact point.

[0078] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the artifact morphology similarity includes:

[0079] In the artifact feature area of ​​each suspected artifact point, the pixel point closest to the vertex position of the upper left corner of the cardiac ultrasound grayscale image is used as the starting point, and the Freeman chain code of the corresponding artifact feature area is obtained by the chain code method; in the suspected artifact point sequence, the DTW distance between the Freeman chain code corresponding to each suspected artifact point and the Freeman chain code corresponding to the previous suspected artifact point is calculated by the dynamic time warping algorithm. It should be noted that the chain code method and the dynamic time warping algorithm are technical means well known to those skilled in the art, and will not be further defined or elaborated here.

[0080] For two artifact feature regions with the same shape and size, the Freeman chain codes at the same starting point are usually the same, and the closer the shapes of the two artifact feature regions are, the more similar the corresponding Freeman chain codes are. The dynamic time warping algorithm can reduce the error when the two Freeman chain codes are compared for similarity, so that the smaller the DTW distance between the two Freeman chain codes corresponding to the two suspected artifact points, the higher the shape and size similarity of the artifact feature regions of the corresponding two suspected artifact points, that is, the more it conforms to the characteristics of the high morphological similarity between the artifact regions, and the greater the corresponding artifact probability. The area difference of the feature region can characterize the similarity of the two artifact feature regions in area. Therefore, the smaller the area difference of the feature region and the smaller the DTW distance, the higher the artifact probability of the corresponding suspected artifact point. Therefore, the product between the DTW distance and the area difference of the feature region is further negatively correlated to determine the feature region similarity of each suspected artifact point; so that the greater the similarity of the feature region, the higher the corresponding artifact probability. Since the artifact credibility can also characterize the corresponding artifact probability, the artifact credibility is further combined with the feature region similarity, and the product between the feature region similarity and the artifact credibility is taken as the artifact morphology similarity of each suspected artifact point.

[0081] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the artifact morphology similarity includes: ;in, For the In the sequence of suspected artifact points of suspected calcified pixels, The similarity of the artifact morphology of the suspected artifact points; For the In the sequence of suspected artifact points of suspected calcified pixels, The credibility of the artifacts of the suspected artifact points; For the In the sequence of suspected artifact points of suspected calcified pixels, The area of ​​the artifact feature region of the suspected artifact point; For the In the sequence of suspected artifact points of suspected calcified pixels, The area of ​​the artifact feature region of the suspected artifact point; For the In the sequence of suspected artifact points of suspected calcified pixels, The difference in the area of ​​the characteristic regions of the suspected artifact points; For the In the suspected artifact point sequence of suspected calcified pixels, the The Freeman chain code corresponding to the suspected artifact point is The DTW distance between the Freeman chain codes corresponding to the suspected artifact points; For the In the sequence of suspected artifact points of suspected calcified pixels, Similarity of feature regions of suspected artifact points; is an exponential function with a natural constant as its base.

[0082] Step S105: determining the grayscale correction weight of each suspected artifact point according to the artifact grayscale gradient and the artifact morphology similarity; correcting the grayscale value of each suspected artifact point according to the grayscale correction weight to determine the enhanced cardiac ultrasound enhanced image.

[0083] Finally, the grayscale correction weight of each suspected artifact point is determined based on the artifact grayscale gradient and artifact morphology similarity that characterize the artifact probability, so that the grayscale value of each suspected artifact point is further corrected according to the grayscale correction weight, thereby reducing the influence of the artifact on the cardiac ultrasound image and making the enhanced cardiac ultrasound enhanced image more accurate, thereby improving the effect of medical teaching based on cardiac ultrasound enhanced images.

[0084] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the grayscale correction weight includes:

[0085] Since the greater the grayscale gradient of the artifact and the greater the similarity of the artifact morphology, the more the corresponding suspected artifact point conforms to the characteristics of the artifact area, and therefore the more its grayscale value needs to be corrected. Therefore, the present application further normalizes the product between the grayscale gradient of the artifact and the similarity of the artifact morphology, and determines the grayscale correction weight of each suspected artifact point, so that the higher the artifact probability of the suspected artifact point, that is, the greater the corresponding grayscale correction weight, the greater the grayscale value of the corresponding suspected artifact point needs to be adjusted.

[0086] Preferably, in some possible implementations of the embodiments of the present invention, the process of acquiring the cardiac ultrasound enhanced image includes:

[0087] Count all suspected artifact points corresponding to all suspected calcified pixel points; considering that the artifact area is produced by the reflection of calcified tissue with higher grayscale value, the grayscale value of the artifact area is usually relatively high, so it is necessary to lower the grayscale value; therefore, the more the corresponding suspected artifact point conforms to the characteristics of the artifact area, that is, the greater the probability of the artifact is, the lower the corresponding grayscale value needs to be adjusted, so it is necessary to perform negative correlation mapping on the grayscale correction weight and then correct the grayscale value, so further in the cardiac ultrasound grayscale image, the product of the negative correlation mapping value of the grayscale correction weight of each suspected artifact point and the corresponding grayscale value is used as the enhanced grayscale value of each suspected artifact point; the grayscale values ​​of all suspected artifact points in the cardiac ultrasound grayscale image are replaced with the corresponding enhanced grayscale values, and the enhanced cardiac ultrasound enhanced image is determined.

[0088] In a specific implementation of the embodiment of the present invention, the process of obtaining the enhanced gray value is expressed by the formula: ;in, For the In the sequence of suspected artifact points of suspected calcified pixels, Enhanced gray value of suspected artifact points; For the In the sequence of suspected artifact points of suspected calcified pixels, Gray value of suspected artifact points; For the In the sequence of suspected artifact points of suspected calcified pixels, The similarity of the artifact morphology of the suspected artifact points; For the In the sequence of suspected artifact points of suspected calcified pixels, The grayscale gradient of the suspected artifact points; For the In the sequence of suspected artifact points of suspected calcified pixels, Grayscale correction weight of suspected artifact points; is a linear normalization function; at this point, the required enhanced cardiac ultrasound enhanced image is acquired.

[0089] In summary, the present application first screens out suspected calcification pixels based on the characteristics of highlighted pixels of tissue calcification, and screens out suspected artifact points that may be affected based on the imaging principle; further determines the artifact feature area of ​​each suspected artifact point based on the grayscale similarity feature based on the artifact area, and determines the possibility of artifacts based on the characteristics that the grayscale attenuation and the law of unchanged artifact area will be destroyed after the artifact feature area is affected by other factors; then, based on the grayscale gradient trend and the characteristics of unchanged morphology of the artifact area, determines the grayscale correction weight that characterizes the degree of influence of the artifact, and corrects the grayscale value of each suspected artifact point according to the grayscale correction weight, thereby reducing the influence of the artifact on the cardiac ultrasound image, solving the problem that the influence of the step artifact cannot be effectively removed by the histogram equalization method alone, resulting in poor effect of image enhancement by histogram equalization, making the acquired enhanced cardiac ultrasound enhanced image more accurate, thereby improving the effect of medical teaching based on cardiac ultrasound enhanced images.

[0090] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0091] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A method for automatically acquiring cardiac ultrasound images, characterized in that: The method comprises: Acquire a cardiac ultrasound grayscale image of the patient through an ultrasound probe; perform threshold segmentation based on the grayscale value distribution in the cardiac ultrasound grayscale image to screen out suspected calcified pixels; determine a corresponding suspected artifact point sequence based on the position of the suspected calcified pixels relative to the corresponding pixel of the ultrasound probe; Determine the reference degree of the artifact region between each pixel point and each suspected artifact point according to the similarity of the grayscale values ​​of all pixels between each pixel point and each suspected artifact point; determine the corresponding artifact feature region according to the reference degree of the artifact region of each pixel point corresponding to each suspected artifact point; In the sequence of suspected artifact points, the anti-interference degree of each suspected artifact point is determined according to the compliance of the grayscale attenuation law between each suspected artifact point and an adjacent suspected artifact point and the similarity of the area of ​​the artifact feature region; the artifact credibility is determined based on the minimum value of the anti-interference degree between each suspected artifact point and an adjacent suspected artifact point; According to the overall grayscale attenuation trend of the artifact feature area between adjacent suspected artifact points combined with the corresponding artifact credibility, the artifact grayscale gradient of all suspected artifact points is determined; according to the morphological similarity of the artifact feature area between adjacent suspected artifact points combined with the corresponding artifact credibility, the artifact morphological similarity of each suspected artifact point is determined; According to the grayscale gradient of the artifact and the similarity of the artifact morphology, a grayscale correction weight of each suspected artifact point is determined; the grayscale value of each suspected artifact point is corrected according to the grayscale correction weight to determine an enhanced cardiac ultrasound enhanced image.

2. The method for automatically acquiring cardiac ultrasound images according to claim 1, characterized in that: The process of obtaining the suspected calcified pixel points includes: The grayscale threshold is determined by the Otsu threshold method according to the grayscale values ​​of all pixels in the cardiac ultrasound grayscale image; and the pixels whose grayscale values ​​are greater than the grayscale threshold are regarded as suspected calcification pixels.

3. The method for automatically acquiring cardiac ultrasound images according to claim 1, characterized in that: The process of acquiring the suspected pseudo-image point sequence includes: The pixel point corresponding to the position of the ultrasound probe is used as the ultrasound probe point; and the artifact interval distance of the suspected calcification pixel point is determined according to the Euclidean distance between the ultrasound probe point and the suspected calcification pixel point; On a ray pointing to the suspected calcified pixel point with the ultrasound probe point as the endpoint, a suspected artifact point is selected at every artifact interval distance; all suspected artifact points are arranged in sequence along the extension direction of the ray to determine a suspected artifact point sequence corresponding to the suspected calcified pixel point.

4. The method for automatically acquiring cardiac ultrasound images according to claim 1, characterized in that: The process of obtaining the reference degree of the artifact area includes: Taking each suspected artifact point as a target artifact point in turn; calculating the grayscale value variance of all pixels on the line between each pixel point in the cardiac ultrasound grayscale image and the target artifact point; performing negative correlation mapping on the difference between the grayscale value of each pixel point in the cardiac ultrasound grayscale image and the grayscale value of the target artifact point, and determining the reference grayscale similarity of each pixel point; The product of the negative correlation mapping value of the gray value variance and the reference gray similarity is normalized to determine the artifact region reference degree between each pixel point and the target artifact point in the cardiac ultrasound gray image.

5. The method for automatically acquiring cardiac ultrasound images according to claim 1, characterized in that: The process of acquiring the artifact feature area includes: Determine a reference degree threshold value by using the Otsu threshold method according to the reference degrees of all artifact regions corresponding to each suspected artifact point; use a pixel point whose artifact region reference degree between each suspected artifact point and the pixel point is greater than the reference degree threshold value as an artifact feature point of each suspected artifact point; The connected domain composed of all the artifact feature points corresponding to each suspected artifact point in the cardiac ultrasound grayscale image is used as the reference feature region of each suspected artifact point; and the reference feature region where each suspected artifact point is located is used as the corresponding artifact feature region.

6. The method for automatically acquiring cardiac ultrasound images according to claim 1, characterized in that: The process of obtaining the anti-interference degree includes: In the suspected artifact point sequence, negative correlation mapping is performed on the difference between the grayscale value of each suspected artifact point and the grayscale value of the previous suspected artifact point to determine the conformity degree of the attenuation law of each suspected artifact point; Determine the area difference of the feature region of each suspected artifact point according to the difference between the area of ​​the artifact feature region of each suspected artifact point and the area of ​​the artifact feature region of the last suspected artifact point; The product of the negative correlation mapping value of the area difference of the characteristic region and the degree of compliance with the attenuation law is normalized to determine the anti-interference degree of each suspected artifact point.

7. The method for automatically acquiring cardiac ultrasound images according to claim 1, characterized in that: The process of obtaining the grayscale gradient of the artifact image includes: The mean grayscale value of all pixels in the artifact feature area of ​​each suspected artifact point is taken as the artifact grayscale feature value of each suspected artifact point; In the sequence of suspected artifact points, the negative correlation mapping value of the difference between the artifact grayscale eigenvalue of each suspected artifact point and the artifact grayscale eigenvalue of the previous suspected artifact point is used as the grayscale gradient eigenvalue of each suspected artifact point; and the product of the artifact credibility and the grayscale gradient eigenvalue is used as the artifact grayscale gradient of each suspected artifact point.

8. The method for automatically acquiring cardiac ultrasound images according to claim 6, characterized in that: The process of obtaining the artifact morphology similarity includes: In the artifact feature area of ​​each suspected artifact point, the pixel point closest to the vertex position of the upper left corner of the cardiac ultrasound grayscale image is taken as the starting point, and the Freeman chain code of the corresponding artifact feature area is obtained by the chain code method; In the suspected artifact point sequence, a DTW distance between a Freeman chain code corresponding to each suspected artifact point and a Freeman chain code corresponding to a previous suspected artifact point is calculated by a dynamic time warping algorithm; The product of the DTW distance and the area difference of the feature region is negatively correlated to determine the feature region similarity of each suspected artifact point; the product of the feature region similarity and the artifact credibility is used as the artifact morphology similarity of each suspected artifact point.

9. The method for automatically acquiring cardiac ultrasound images according to claim 1, characterized in that: The grayscale correction weight acquisition process includes: The product of the grayscale gradient of the artifact and the morphological similarity of the artifact is normalized to determine the grayscale correction weight of each suspected artifact point.

10. The method for automatically acquiring cardiac ultrasound images according to claim 1, characterized in that: The process of acquiring the cardiac ultrasound enhanced image includes: Count all suspected artifact points corresponding to all suspected calcified pixel points; in the cardiac ultrasound grayscale image, take the product of the negative correlation mapping value of the grayscale correction weight of each suspected artifact point and the corresponding grayscale value as the enhanced grayscale value of each suspected artifact point; replace the grayscale values ​​of all suspected artifact points in the cardiac ultrasound grayscale image with the corresponding enhanced grayscale values, and determine the enhanced cardiac ultrasound enhanced image.

Images