CN112651984A

CN112651984A - Blood vessel lumen intimal contour extraction method and device, ultrasonic equipment and storage medium

Info

Publication number: CN112651984A
Application number: CN202011640073.9A
Authority: CN
Inventors: 胡浩晖; 黎英云; 朱彦聪
Original assignee: Sonoscape Medical Corp
Current assignee: Sonoscape Medical Corp
Priority date: 2020-12-31
Filing date: 2020-12-31
Publication date: 2021-04-13

Abstract

The application discloses a method and a device for extracting a blood vessel lumen intima contour, an ultrasonic device and a computer readable storage medium, wherein the method comprises the following steps: acquiring a first image of the cross section in the blood vessel, and calculating the similarity between the first image and a second image of the determined lumen intima contour line; if the similarity is larger than or equal to a first preset value, calculating the optimal torsion angle of the first image and the second image; determining an initial lumen intima contour line of the first image according to the lumen intima contour line of the second image and the optimal torsion angle; and fitting the initial lumen intima contour line based on the initial lumen intima contour line to obtain a final lumen intima contour line of the first image. Therefore, the method for extracting the intima contour of the blood vessel lumen improves the speed and the accuracy of extracting the intima contour of the blood vessel lumen.

Description

Blood vessel lumen intimal contour extraction method and device, ultrasonic equipment and storage medium

Technical Field

The present application relates to the field of ultrasound image technology, and more particularly, to a method and an apparatus for extracting a contour of a vessel lumen intima, an ultrasound device, and a computer-readable storage medium.

Background

Intravascular ultrasound (IVUS) refers to a medical imaging technique using a special catheter with an ultrasound probe connected to the end, in combination with non-invasive ultrasound and invasive catheter techniques. The intima-adventitia edge of the coronary artery vessel wall is an important basis for diagnosing and quantitatively analyzing various parameters of coronary artery diseases, so that the accurate detection of the intima-adventitia edge in the IVUS image has great significance for clinical diagnosis and treatment of the coronary artery diseases.

In the related art, the IVUS image is twisted along with the beating of the heart, so that the extraction of the lumen intima contour is not facilitated, and the extraction speed and the accuracy of the blood vessel lumen intima contour are low. Therefore, how to improve the accuracy of the blood vessel lumen intima contour extraction is a technical problem to be solved by those skilled in the art.

Disclosure of Invention

The application aims to provide a method and a device for extracting a blood vessel lumen intima contour, ultrasonic equipment and a computer-readable storage medium, and the accuracy of the blood vessel lumen intima contour extraction is improved.

In order to achieve the above object, the present application provides a method for extracting a contour of an intima of a blood vessel lumen, comprising:

acquiring a first image of an intravascular cross section, and calculating the similarity between the first image and a second image of the determined lumen intima contour line;

if the similarity is larger than or equal to a first preset value, calculating the optimal torsion angle of the first image and the second image;

determining an initial lumen intimal contour line of the first image according to the lumen intimal contour line of the second image and the optimal torsion angle;

and fitting the initial lumen intima contour line based on the initial lumen intima contour line to obtain the final lumen intima contour line of the first image.

Wherein calculating a similarity between the first image and a second image of the determined intima contour comprises:

determining a coordinate range corresponding to the region of interest based on the lumen intimal contour line of the second image;

extracting a first region of interest in the second image according to the coordinate range, and extracting a second region of interest in the first image according to the coordinate range;

calculating a similarity between the first region of interest and the second region of interest as a similarity between the first image and the second image.

Wherein, the coordinate range corresponding to the region of interest is determined based on the lumen intima contour line of the second image, and the method comprises the following steps:

calculating the maximum distance between the contour point in the lumen intima contour line of the second image and the image center of the second image, and taking the sum of the maximum distance and a preset offset distance as the length of the target side;

determining a target square area, and taking the coordinate range of the target square area as the coordinate range corresponding to the region of interest; and the center of the target square area is the image center, and the side length is the target side length.

Wherein calculating an optimal torsion angle of the first image and the second image comprises:

determining an initial torsion angle range and an initial angle interval, and determining a target angle based on the initial angle interval in the initial torsion angle range;

calculating the similarity between the first image and the second image under each target angle, and determining the maximum similarity in all the similarities;

judging whether a preset condition is met; the preset condition comprises that the maximum similarity is greater than or equal to a second preset value, or the updating times are greater than or equal to preset times;

if so, taking the target angle corresponding to the maximum similarity as the optimal torsion angle of the first image and the second image;

if not, updating the initial torsion angle range and the initial angle interval, increasing the updating times by one, re-determining a target angle based on the updated angle interval in the updated torsion angle range, and re-entering the step of calculating the similarity between the first image and the second image at each target angle; the updated torsion angle range is smaller than the torsion angle range before updating, the updated angle interval is smaller than the angle interval before updating, and the number of the target angles before updating is the same as that of the target angles after updating.

Wherein determining the initial intraluminal contour of the first image according to the intraluminal contour of the second image and the optimal torsion angle comprises:

calculating the coordinates of the contour points in the first image by using a contour point coordinate calculation formula, and determining the connecting lines of the contour points as initial lumen intima contour lines; the contour point coordinate calculation formula specifically includes:

wherein the coordinates of the contour point in the first image are (x ', y'), the coordinates of the lumen intima contour line in the second image are (x, y), and the image center coordinates of the first image are (x, y)₀,y₀) And A is the optimal torsion angle, and theta is an included angle between a connecting line of the contour point (X, y) and the image center of the first image and the positive direction of the X axis.

Wherein, still include:

if the similarity is smaller than a first preset value, determining a lumen wall area in the first image, and determining an initial lumen intima contour line corresponding to the lumen wall area;

and fitting the initial lumen intima contour line to obtain a final lumen intima contour line of the first image.

Wherein, it is right to fit the initial lumen intima contour line, obtain the final lumen intima contour line of the first image, include:

and fitting the initial lumen intima contour line by using a vector field convolution active contour model to obtain a final lumen intima contour line of the first image.

To achieve the above object, the present application provides a blood vessel lumen intima contour extraction device, comprising:

the acquisition module is used for acquiring a first image of the cross section in the blood vessel and calculating the similarity between the first image and a second image of the determined lumen intima contour line; if the similarity is larger than or equal to a first preset value, starting a working process of a calculation module;

the calculation module is used for calculating the optimal torsion angle of the first image and the second image;

a first determining module, configured to determine an initial intraluminal contour line of the first image according to the intraluminal contour line of the second image and the optimal torsion angle;

and the fitting module is used for fitting the initial lumen intima contour line based on the initial lumen intima contour line to obtain the final lumen intima contour line of the first image.

To achieve the above object, the present application provides an ultrasound apparatus comprising:

a memory for storing a computer program;

a processor for implementing the steps of the blood vessel lumen intima contour extraction method when executing the computer program;

a display for displaying the intravascular ultrasound image and the intimal contour line in the intravascular ultrasound image.

To achieve the above object, the present application provides a computer-readable storage medium having stored thereon a computer program, which when executed by a processor, implements the steps of the vessel lumen intima contour extraction method as described above.

According to the scheme, the method for extracting the intima contour of the blood vessel lumen comprises the following steps: acquiring a first image of an intravascular cross section, and calculating the similarity between the first image and a second image of the determined lumen intima contour line; if the similarity is larger than or equal to a first preset value, calculating the optimal torsion angle of the first image and the second image; determining an initial lumen intimal contour line of the first image according to the lumen intimal contour line of the second image and the optimal torsion angle; and fitting the initial lumen intima contour line based on the initial lumen intima contour line to obtain the final lumen intima contour line of the first image.

According to the method for extracting the outline of the intima of the blood vessel lumen, when the similarity between the second image and the first image is larger than or equal to the first preset value, the initial intima contour line of the lumen of the first image is determined based on the intima contour line of the lumen of the second image, clustering operation does not need to be carried out on pixel points in the first image again, the speed of determining the initial intima contour line of the lumen of the first image is improved, and then the extraction speed of the intima contour of the blood vessel lumen is improved. Otherwise, the initial lumen intima contour line of the first image is determined by other methods, so that the problem of inaccurate fitting caused by large image change is avoided, and the accuracy of the extraction of the lumen intima contour line of the blood vessel is improved. Meanwhile, when the similarity is larger than or equal to the first preset value, the optimal torsion angle between the second image and the first image is calculated, the lumen intima contour line of the second image and the optimal torsion angle are used for determining the initial lumen intima contour line of the first image, the problem of IVUS image torsion caused by heart beating is solved, and the accuracy of blood vessel lumen intima contour extraction is further improved. Therefore, according to the method for extracting the blood vessel lumen intima contour, when the similarity between the second image and the first image is larger than or equal to the first preset value, the initial lumen intima contour line of the first image is determined by using the lumen intima contour line of the second image and the optimal torsion angle, the image processing flow is simplified, and the speed and the accuracy of extracting the blood vessel lumen intima contour are improved. The application also discloses a blood vessel lumen intimal contour extraction device, ultrasonic equipment and a computer readable storage medium, which can also realize the technical effects.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts. The accompanying drawings, which are included to provide a further understanding of the disclosure and are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure without limiting the disclosure. In the drawings:

FIG. 1 is a flow diagram illustrating a method for vessel lumen intimal contour extraction in accordance with an exemplary embodiment;

FIG. 2a is a schematic diagram illustrating an ellipse after fitting according to an exemplary embodiment;

FIG. 2b is a schematic diagram illustrating an edge image according to an exemplary embodiment;

FIG. 2c is a schematic diagram illustrating a final endoluminal contour according to an exemplary embodiment;

FIGS. 3a and 3b are schematic diagrams illustrating before and after removal of a catheter effect according to an exemplary embodiment;

FIG. 4a is a schematic diagram illustrating morphological processing according to an exemplary embodiment;

FIG. 4b is a schematic diagram illustrating fuzzy clustering in accordance with an exemplary embodiment;

FIG. 4c is a schematic diagram illustrating a threshold segmentation in accordance with an exemplary embodiment;

FIG. 5 is a schematic illustration of a connected region shown in accordance with an exemplary embodiment;

FIG. 6a is a schematic diagram illustrating removal of an outer candidate lumen wall region in accordance with an exemplary embodiment;

FIG. 6b is a schematic diagram of a first polar image shown in accordance with an exemplary embodiment;

FIG. 6c is a schematic diagram illustrating removal of an inner candidate lumen wall region in accordance with an exemplary embodiment;

FIG. 7a is a schematic diagram illustrating a second polar image according to an exemplary embodiment;

FIG. 7b is a schematic diagram illustrating contour points in an initial intimal contour in accordance with an exemplary embodiment;

FIG. 7c is a schematic diagram illustrating interpolation of contour points in accordance with an exemplary embodiment;

FIG. 7d is a schematic diagram illustrating an initial intimal contour line in accordance with an exemplary embodiment;

FIG. 8 is a detailed flowchart of step S102 in FIG. 1;

FIG. 9 is a flow chart illustrating another method of vessel lumen intimal contour extraction in accordance with an exemplary embodiment;

FIGS. 10a and 10b are schematic views of a first region of interest and a second region of interest shown in accordance with an exemplary embodiment;

FIG. 11 is a block diagram illustrating a vascular intraluminal lining contour extraction device in accordance with an exemplary embodiment;

FIG. 12 is a block diagram of an ultrasound device shown in accordance with an exemplary embodiment.

Detailed Description

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. It is to be understood that the embodiments described are only a few embodiments of the present application and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application. In addition, in the embodiments of the present application, "first", "second", and the like are used for distinguishing similar objects, and are not necessarily used for describing a specific order or a sequential order.

The embodiment of the application discloses a method for extracting the intima contour of a blood vessel lumen, which improves the accuracy of extracting the intima contour of the blood vessel lumen.

Referring to fig. 1, a flowchart of a method for extracting a contour of an intraluminal lining of a blood vessel according to an exemplary embodiment is shown, as shown in fig. 1, including:

s101: acquiring a first image of an intravascular cross section, and calculating the similarity between the first image and a second image of the determined lumen intima contour line;

the execution body of the application can be an ultrasonic device, and the purpose is to extract the lumen intima of the blood vessel in the intravascular ultrasonic image. In this embodiment, the first image and the second image are both IVUS images, the determination mode of the initial intraluminal contour line of the first image is determined according to the similarity between the first image and the second image, and the second image is an image of the determined intraluminal contour line. As a possible implementation manner, a normalized cross-correlation matching method may be used for calculation, and the specific calculation formula is:

where ρ (x, y) is the similarity between the first image and the second image, m is the length of the first image and the second image, n is the width of the first image and the second image, I_x(i, j) is the gray value with coordinates (i, j) in the first image,

is the mean value of the gray levels of the first image, I_y(i, j) is the gray value with coordinates (i, j) in the second image,

is the gray scale average of the second image.

And when the similarity meets a preset condition, determining the initial lumen intima contour line of the first image based on the lumen intima contour line of the second image. Otherwise, the initial lumen intima contour line of the first image is determined by other methods, so that the problem of inaccurate fitting caused by large image change is avoided, and the accuracy of the extraction of the lumen intima contour line of the blood vessel is improved.

As a possible implementation manner, the second image is a previous frame image of the first image, it should be noted that the second image may be a previous frame image of the first image, and may also be a few previous frames of the first image, and in order to ensure that the similarity satisfies the preset condition as much as possible, the number of interval frames between the first image and the second image is less than the preset number of frames.

S102: if the similarity is larger than or equal to a first preset value, calculating the optimal torsion angle of the first image and the second image;

s103: determining an initial lumen intimal contour line of the first image according to the lumen intimal contour line of the second image and the optimal torsion angle;

in specific implementation, if the similarity is greater than or equal to the first preset value, the optimal torsion angle between the second image and the first image is calculated, the lumen intima contour line of the second image and the optimal torsion angle are used for determining the initial lumen intima contour line of the first image, the problem of IVUS image torsion caused by heart beating is solved, and the accuracy of blood vessel lumen intima contour extraction is further improved.

As a possible embodiment, the step of determining the initial intraluminal contour of the first image from the intraluminal contour of the second image and the optimal torsion angle includes: calculating the coordinates of the contour points in the first image by using a contour point coordinate calculation formula, and determining the connecting lines of the contour points as initial lumen intima contour lines; the contour point coordinate calculation formula specifically includes:

S104: and fitting the initial lumen intima contour line based on the initial lumen intima contour line to obtain the final lumen intima contour line of the first image.

Because the initial lumen intima contour line deviates from the actual blood vessel lumen, in the step, the initial lumen intima contour line is subjected to evolution fitting, and the final lumen intima contour line is extracted and displayed in the intravascular ultrasound image.

As a possible implementation, the step may include: and fitting the initial lumen intima contour line by using a vector field convolution active contour model to obtain a final lumen intima contour line of the first image. In a specific implementation, an ellipse fitting is performed on the initial intima contour line to reduce the complexity of contour line fitting calculation, and as shown in fig. 2a, the initial intima contour line v(s) ═ x(s), y (s)) can be expressed as:

wherein (x)_c,y_c) And r and theta are the distance radius and the angle between the contour point in the initial intima contour line and the image center position respectively.

The contour line in the contour model satisfies the following equation:

wherein E_intThe initial intima contour line external force field comprises two external force field parameters alpha and beta, wherein the parameter alpha controls contour line tension, the curve is more difficult to stretch when the value of the parameter alpha is larger, the curve shrinks more quickly, the parameter beta controls curve rigidity, the curve is more difficult to bend when the value of the parameter beta is larger, and the curve is smoother. And E_ext[v(s)]F, convolved with the image edge by a vector field convolution kernel, as an external force term for an image feature_vfcTo represent F_vfc(x,y)＝f(x,y)*K(x,y)＝[f(x,y)*u(x,y),f(x,y)*v(x,y)]Where f (x, y) is the edge image detected by phase consistency of the ultrasound image of the cross-section in the blood vessel, as shown in fig. 4b, K (x, y) is a vector field convolution kernel defined as K (x, y) ═ u (x, y), v (x, y)]M (x, y) n (x, y), u (x, y), v (x, y) are the vector field convolution kernel components in the horizontal and vertical methods, respectively, while n (x, y) points to the unit vector at the kernel origin (0,0), and m (x, y) is the modulus value of each vector. Minimizing the above energy equation E_SnakeThen the following euler formula needs to be satisfied:

if v(s) is taken as a function of time, it can be expressed as v (s, t), so that an iterative function with v (s, t) as a variable can be obtained by minimizing the above formula, and an optimal intraluminal contour can be found through multiple iterations, as shown in fig. 4 c. The iterative function with v (s, t) as a variable is:

according to the method for extracting the blood vessel lumen intima contour, when the similarity between the second image and the first image is larger than or equal to the first preset value, the initial lumen intima contour line of the first image is determined based on the lumen intima contour line of the second image, clustering operation does not need to be carried out on pixel points in the first image again, the speed of determining the initial lumen intima contour line of the first image is improved, and then the extraction speed of the blood vessel lumen intima contour is improved. Otherwise, the initial lumen intima contour line of the first image is determined by other methods, so that the problem of inaccurate fitting caused by large image change is avoided, and the accuracy of the extraction of the lumen intima contour line of the blood vessel is improved. Meanwhile, when the similarity is larger than or equal to the first preset value, the optimal torsion angle between the second image and the first image is calculated, the lumen intima contour line of the second image and the optimal torsion angle are used for determining the initial lumen intima contour line of the first image, the problem of IVUS image torsion caused by heart beating is solved, and the accuracy of blood vessel lumen intima contour extraction is further improved. Therefore, according to the method for extracting the blood vessel lumen intima contour provided by the embodiment of the application, when the similarity between the second image and the first image is greater than or equal to the first preset value, the lumen intima contour line of the second image and the optimal torsion angle are used for determining the initial lumen intima contour line of the first image, the image processing flow is simplified, and the speed and the accuracy of extracting the blood vessel lumen intima contour are improved.

On the basis of the above embodiment, as a preferred implementation, the method further includes: if the similarity is smaller than a first preset value, determining a lumen wall area in the first image, and determining an initial lumen intima contour line corresponding to the lumen wall area; and fitting the initial lumen intima contour line to obtain a final lumen intima contour line of the first image.

In specific implementation, if the similarity is smaller than a first preset value, the first image is preprocessed to obtain a target image.

As a possible implementation, the preprocessing of the ultrasound image may include: and performing down-sampling and filtering processing on the ultrasonic image.

In a specific implementation, due to the large size of the ultrasound image of the intravascular cross section, it needs to be down-sampled, and the down-sampling process can be used to increase the speed of subsequent image processing. For example, a sampling rate of 0.5 is set, i.e., one data is taken out of every two data in the horizontal and vertical directions, so that the image size is reduced by one time. And then, filtering the down-sampled image to eliminate the influence of partial noise on the extraction of the inner membrane edge. For example, the Filter coefficient may be set to [1,4,6,4,1]/16, and during the filtering process, the horizontal convolution operation may be performed on the original image data, and then the vertical convolution operation may be performed on the result to complete the filtering process.

Preprocessing the ultrasound image may further include: catheter effects in the image are removed. It can be understood that because the center of the image of the ultrasound image of the intravascular cross section is the catheter, and ring halo artifacts are near the center of the image, which easily interfere with the subsequent contour extraction, the highlight ring halo artifacts near the catheter can be removed in the preprocessing process. As a possible implementation, the removing the duct effect in the image includes: determining pixel points with the first gray value not being zero in each direction by taking the central position of the image as a starting point as target positions; calculating an average value of the distance between each target position and the image center position, and calculating the sum of the average value and a preset offset distance as a superposition distance; and setting all gray values of pixel points with the distance from the central position of the image to be smaller than the superposition distance as a lower limit value to obtain a second intermediate image. In this embodiment, a pixel point with a first gray value not being zero is searched from the image center position to the 360-degree direction of the circumference as a target position, an average value of distances between each target position of the 360-degree circumference and the image center position is calculated, and then a sum of the average value and a preset offset distance is determined as a superimposition distance, where the preset offset distance is set empirically, and may be set to 15 pixels, for example. And finally, setting all the gray values of the pixel points which are smaller than the superposition distance from the central position of the image as a lower limit value, for example, setting all the gray values as zero. The catheter effect is removed as shown in figure 3a before and in figure 3b after.

Preprocessing the ultrasound image may further include: and carrying out significance enhancement on the interested region in the image so as to improve the gray value of the image of the tissues near the blood vessel wall to highlight the contour of the lumen intima. As a possible implementation, the salient enhancement of the region of interest in the image includes: determining a region of interest in the image; the gray value of each pixel point in the region of interest is greater than a first gray threshold value; calculating the pixel average of all pixel points in the region of interestDetermining a value and a mapping curve based on the pixel average value; and processing the gray value of each pixel point in the image by using the mapping curve. It will be appreciated that the purpose of the significant enhancement is to highlight regions of interest in the tissue near the vessel wall, suppressing regions of non-interest within the vessel lumen. In this embodiment, the number of each gray value in the image is counted to calculate the pixel average value of all the pixel points in the region of interest, and a mapping curve y is established as k × log₂And x, resetting the gray value of each pixel point in the image by using the mapping curve, wherein x is the original gray value of the pixel point, y is the updated gray value of the pixel point, and k is the average value of the pixels. Meanwhile, the gray value of the highlight area is calculated, a mapping curve is established, and finally the original IVUS is subjected to significance enhancement through the mapping curve.

Further, the preprocessing the ultrasound image may further include: and carrying out normalization processing on the gray values of the pixel points in the image. In specific implementation, the gray value of a pixel point in an image is normalized to obtain a target image, and the gray value is prevented from exceeding the interval of 0-255.

It should be noted that the sequence relationship among the several steps of the preprocessing is not strictly limited in this embodiment, and the number of times each step is executed is not strictly limited, and the preprocessing can be flexibly adjusted according to the imaging condition of the actual ultrasound image. As a preferred embodiment, the down-sampling and filtering processing is performed on the ultrasound image, the catheter effect in the image is removed, the region of interest in the image is enhanced in significance, and the normalization processing is performed on the gray value of the pixel point in the image.

Secondly, determining a lumen wall area in the target image, and taking the contour line of the lumen wall area as an initial intima contour line. As a preferred embodiment, the determining the lumen wall region in the target image includes: determining the category of each pixel point in the target image so as to divide the target image into different clustering blocks; and determining a lumen wall area in the clustering block, and removing external noise and internal noise of the lumen wall area. In the specific implementation, a clustering algorithm is utilized to cluster pixel points in a target image, the target image is divided into different clustering blocks, then a lumen wall area is determined in all the clustering blocks, and other clustering blocks are taken as noise to be removed.

Preferably, the determining the category of each pixel point in the target image to segment the target image into different clustering blocks includes: and carrying out clustering operation on pixel points in the target image by a fuzzy C-means clustering algorithm based on a gray level histogram so as to divide the target image into different clustering blocks. It can be understood that, in the fuzzy C-means clustering algorithm based on the gray level histogram, the clustering centers are different gray levels, so that the target image is divided into continuous classes in gray level and position, and then different clustering blocks for segmenting the vascular lumen wall are further extracted.

Specifically, the fuzzy C-means clustering algorithm based on the gray histogram performs clustering operation on the pixel points in the target image to segment the target image into different clustering blocks, including: carrying out corrosion reconstruction on the target image by using morphological processing to obtain an intermediate image, and determining a clustering parameter; the clustering parameters comprise the number of clustering centers, kernel processing scales and fuzzy factors; determining different gray values as initial clustering centers according to the range of the gray level histogram of the intermediate image and the number of the clustering centers, and determining a gray value membership function based on the fuzzy factor and Euclidean clustering of the gray level values in the gray level histogram and the initial clustering centers; updating the initial clustering center and the gray value membership function through iteration until the fuzzy C mean value objective function is optimized or the iteration times reach preset times; and determining the maximum value in the updated clustering center as a second gray threshold, and setting all gray values of pixel points with gray values larger than or equal to the second gray threshold in the intermediate image as an upper limit value and gray values of pixel points with gray values smaller than the second gray threshold in the intermediate image as a lower limit value so as to divide the target image into different clustering blocks.

In a specific implementation, since speckle noise exists in the target image, it is necessary to first perform erosion reconstruction using morphological processing, as shown in fig. 4 a. In the fuzzy C-means clustering process based on the gray level histogram, firstly, clustering parameters including the number of clustering centers, kernel processing scale and fuzzy factors are determined. Different gray values can be determined according to the range of the gray histogram of the image and the number of the clustering centers, and the gray values are used as initial clustering centers. Further, a gray value membership function can be deduced according to a membership function matrix in a standard fuzzy C-means clustering algorithm:

wherein C is the number of cluster centers, k is 1, 2, 3, and C, m is a blurring factor, d is_ij＝||x_i-v_iAnd | | is the Euclidean distance from the gray value of the jth pixel point to the ith clustering center.

Secondly, updating the initial clustering center and the gray value membership function through loop iteration to optimize the fuzzy C-means objective function or terminate the loop iteration when the iteration number reaches a preset number, and finally obtaining the final gray value membership function and the clustering center, wherein the calculation formula is as follows:

clustering center:

an objective function:

wherein v is_iIs the ith cluster center, x_iIs the gray value of the jth pixel point, and n is the total number of pixel points in the image. And determining the category of each pixel point in the image according to the final grey value membership function, and dividing the image into different parts, as shown in fig. 4 b.

Next, the maximum value in the final clustering center is determined as a second gray threshold, the gray value of the pixel whose gray value is greater than or equal to the second gray threshold is set to 255, and the gray value of the pixel whose gray value is less than the second gray threshold is set to 0, so as to segment the blood vessel lumen region, as shown in fig. 4 c.

It can be understood that the image after fuzzy C-means clustering is divided into different regions, and the different image regions are identified by an image connectivity method, as shown in fig. 5. Determining the lumen wall area according to the characteristics of the lumen wall area closest to the center of the image, determining the area center position of each candidate lumen wall area, and determining the candidate lumen wall area with the closest distance between the area center position and the image center position as the final lumen wall area.

In order to eliminate the influence of other interference-like blocks such as pericardium and artifact, it is necessary to remove other clustering blocks except the final lumen wall region, i.e. to remove external noise and internal noise of the lumen wall region.

It can be understood that since the external noise is far away from the final lumen wall region, the euclidean distance between the region center position of the cluster block and the region center position of the final lumen wall region can be calculated under the rectangular coordinate system for removal. The internal noise is mostly caused by factors such as guide wire artifacts and strong reflection of calcified plaques, is located in the final lumen wall area, and cannot be removed in the above way, so that the image can be subjected to polar coordinate conversion, and the internal noise can be removed. Of course, the external noise can be removed in the polar coordinate image, but the external noise and the internal noise are removed in the rectangular coordinate image and the polar coordinate image respectively, so that confusion between the external noise and the internal noise can be avoided, and the removal effect is better.

As a possible embodiment, removing external noise from the lumen wall region includes: calculating Euclidean distances between the region center positions of all the clustering blocks and the region center position of the lumen wall region; counting the overlapping area between the surrounding rectangles corresponding to all the clustering blocks and the surrounding rectangles corresponding to the lumen wall area; and determining the clustering block with the Euclidean distance larger than a preset distance threshold value and the overlapping area smaller than a first preset area threshold value as the external noise of the lumen wall area, and removing the external noise.

In the present embodiment, the removal is performed by presetting a distance threshold and a region overlapping area threshold. Firstly, calculating the Euclidean distance between the final region center position of the lumen wall region and the region center positions of all the clustering blocks, determining the surrounding rectangles corresponding to the clustering blocks, counting the overlapping area between the surrounding rectangle corresponding to each clustering block and the final surrounding rectangle corresponding to the lumen wall region, when the overlapping area is smaller than a first preset area threshold value and the Euclidean distance is larger than a preset distance threshold value, defining the clustering block as an external candidate lumen wall region, and removing the external noise, as shown in FIG. 6 a.

As a possible embodiment, removing internal noise of the lumen wall region includes: performing polar coordinate conversion on the image without the external noise to obtain a first polar coordinate image; determining the area of each clustering block, the gray average value of pixel points and the number of overlapped pixel points with the lumen wall area in the first polar coordinate image; and determining the clustering blocks with the area smaller than a second preset area threshold, the gray average value larger than a third gray threshold and the number of overlapped pixel points larger than a preset number as the internal noise of the tube cavity wall area, and removing the internal noise.

The image from which the external noise is removed is subjected to polar coordinate conversion to obtain a first polar coordinate image, as shown in fig. 6 b. The rectangular coordinates (x, y) and polar coordinates (r, θ) are transformed as follows:

in the first polar coordinate image, calculating the area of the candidate lumen wall area and the gray average value of the pixel points, and determining the inner candidate lumen wall area by combining the number of the overlapped pixel points of the final lumen wall area. In a specific implementation, the final lumen wall region and all candidate lumen wall regions may be projected onto the abscissa of the polar coordinate image, and the number of overlapping pixel points is determined according to the overlapping length of the projection corresponding to each candidate lumen wall region and the projection corresponding to the final lumen wall region. When the area of the region is smaller than the second preset area threshold, the average gray level is larger than the third gray level threshold, and the number of overlapped pixel points is larger than the preset number, the candidate lumen wall region is defined as an internal candidate lumen wall region, and the internal candidate lumen wall region is removed from the rectangular coordinate image, as shown in fig. 6 c.

It should be noted that due to the complexity of coronary vessels, the shape of the imaged IVUS image varies, and there is a lot of noise and artifact interference, so that the clustering parameters need to be adjusted according to different tissue shapes. In this embodiment, after the final lumen wall region is extracted, the maximum lateral width thereof is determined. And judging whether the ratio of the transverse maximum width to the width of the sixth intermediate image is greater than a preset value, if so, indicating that the fuzzy clustering effect is good, and better segmenting the lumen area of the blood vessel so as to facilitate the initial positioning of the subsequent intima contour. If not, the clustering parameters need to be updated, and the operation is repeated. The preset value is not limited herein, and may be set to 0.7, for example.

Finally, the image from which the external noise and the internal noise are removed is subjected to polar coordinate conversion to obtain a second polar coordinate image, as shown in fig. 7a, and an edge map of the lumen region is detected through phase consistency. Because the lumen intima contour line is located on the upper part of the image after the polar coordinate change, the pixel point of each column of the second polar coordinate image with the first gray value not being zero is taken as the contour point in the initial intima contour line, as shown in fig. 7 b. The missing part, i.e. the part with zero ordinate, is interpolated cubic by spline to find the points where the edges of the entire intima are smooth, as shown in fig. 7 c. And finally, performing coordinate transformation on the contour points obtained by calculation, and outputting the initial intima contour line in a line segment connection mode, as shown in fig. 7 d. And fitting the initial lumen intima contour line by using the vector field convolution active contour model to obtain the final lumen intima contour line of the first image, which is not described herein any more.

Therefore, in the embodiment, the intima initial contour line is searched based on the fuzzy C-means clustering of the gray histogram, and the vector field convolution active contour model accurately fits the intima of the intravascular ultrasound image according to the initial intima contour line, so that the accuracy of the extraction of the intima contour of the blood vessel lumen is improved.

An embodiment of calculating the optimal torsion angle is described below, that is, as shown in fig. 8, step S102 in the above embodiment may include:

s1021: determining an initial torsion angle range and an initial angle interval, and determining a target angle based on the initial angle interval in the initial torsion angle range;

in a specific implementation, an initial twist angle range and an initial angle interval are determined, and a target angle is determined based on the initial angle interval within the initial twist angle range. For example, if the initial twist angle ranges from-45 degrees to 45 degrees and the initial angle interval is 9 degrees, the target angles are-45 degrees, -36 degrees, -27 degrees, -18 degrees, -9 degrees, 0 degrees, 9 degrees, 18 degrees, 27 degrees, 36 degrees, and 45 degrees, respectively.

S1022: calculating the similarity between the first image and the second image under each target angle, and determining the maximum similarity in all the similarities;

s1023: judging whether a preset condition is met; the preset condition comprises that the maximum similarity is greater than or equal to a second preset value, or the updating times are greater than or equal to preset times; if yes, entering S1024; if not, the process goes to S1025;

s1024: taking the target angle corresponding to the maximum similarity as the optimal torsion angle of the first image and the second image;

s1025: updating the initial torsion angle range and the initial angle interval, increasing the number of updating by one, re-determining the target angle based on the updated angle interval in the updated torsion angle range, and step S1022; the updated torsion angle range is smaller than the torsion angle range before updating, the updated angle interval is smaller than the angle interval before updating, and the number of the target angles before updating is the same as that of the target angles after updating.

In a specific implementation, the similarity between the first image and the second image at each target angle is calculated, the maximum similarity is determined, and if the maximum similarity is greater than or equal to a second preset value or the update frequency is greater than or equal to a preset frequency, the target angle corresponding to the maximum similarity is taken as the optimal torsion angle, where it can be understood that the second preset value is greater than the first preset value. Otherwise, obtaining an updated torsion angle range according to the initial torsion angle range based on the angle range of the target angle corresponding to the maximum similarity, and re-determining the angle interval in the updated torsion angle range. It can be understood that the range of the torsion angle after updating is smaller than the range of the torsion angle before updating, the angle interval after updating is smaller than the angle interval before updating, and the number of the target angles before updating is the same as the number of the target angles after updating, so as to ensure the selection precision of the optimal torsion angle. In the above-mentioned example, if the target angle corresponding to the maximum similarity is 18 degrees, the torsion angle range may be updated to 9 degrees to 27 degrees, the angle interval may be updated to 1.8 degrees, and the updated target angles may be 9 degrees, 10.8 degrees, 12.6 degrees, 14.4 degrees, 16.2 degrees, 18 degrees, 19.8 degrees, 21.6 degrees, 23.4 degrees, 25.2 degrees, and 27 degrees, respectively.

Therefore, in the embodiment, the optimal torsion angle is calculated by continuously updating the torsion angle range and the angle interval, the similarity between the first image and the second image at each angle does not need to be calculated, and the calculation amount for calculating the optimal torsion angle is reduced.

The embodiment of the application discloses a method for extracting the intimal contour of a blood vessel lumen, and compared with the previous embodiment, the embodiment further explains and optimizes the technical scheme. Specifically, the method comprises the following steps:

referring to fig. 9, a flowchart of another vessel lumen intimal contour extraction method is shown according to an exemplary embodiment, as shown in fig. 9, including:

s201: acquiring a first image of the cross section in the blood vessel, and determining a coordinate range corresponding to the region of interest based on the lumen intima contour line of the second image;

s202: extracting a first region of interest in the second image according to the coordinate range, and extracting a second region of interest in the first image according to the coordinate range;

s203: calculating a similarity between the first region of interest and the second region of interest as a similarity between the first image and the second image.

In the present embodiment, in order to reduce the amount of calculation in calculating the similarity, the second region of interest and the first region of interest are extracted in the first image and the second image, respectively. When the similarity is calculated, only the similarity between corresponding pixel points in the first interested region and the second interested region needs to be calculated.

When extracting the region of interest, it is necessary to determine a coordinate range corresponding to the region of interest based on the endocardial contour of the second image, and as a feasible implementation manner, this step may include: calculating the maximum distance between the contour point in the lumen intima contour line of the second image and the image center of the second image, and taking the sum of the maximum distance and a preset offset distance as the length of the target side; determining a target square area, and taking the coordinate range of the target square area as the coordinate range corresponding to the region of interest; and the center of the target square area is the image center, and the side length is the target side length. In specific implementation, the maximum distance between each contour point in the lumen intima contour line of the second image and the center of the image is calculated, and the sum of the maximum distance and a preset offset distance is taken as the side length of the target square region to obtain a coordinate range corresponding to the region of interest. The first region of interest is shown in fig. 10a and the second region of interest is shown in fig. 10 b.

S204: if the similarity is larger than or equal to a first preset value, calculating the optimal torsion angle of the first image and the second image;

s205: determining an initial lumen intimal contour line of the first image according to the lumen intimal contour line of the second image and the optimal torsion angle;

s206: and fitting the initial lumen intima contour line based on the initial lumen intima contour line to obtain the final lumen intima contour line of the first image.

Therefore, in the embodiment, the second region of interest and the first region of interest are extracted from the first image and the second image respectively, and the similarity between the first region of interest and the second region of interest is calculated as the similarity between the first image and the second image, so that the calculation amount in calculating the similarity is reduced, and the extraction speed of the lumen intima contour line is further improved.

In the following, a blood vessel lumen intima contour extraction device provided by an embodiment of the present application is introduced, and a blood vessel lumen intima contour extraction device described below and a blood vessel lumen intima contour extraction method described above may be referred to each other.

Referring to fig. 11, a block diagram of a blood vessel lumen intima contour extraction device according to an exemplary embodiment is shown, as shown in fig. 11, including:

an obtaining module 100, configured to obtain a first image of an intravascular cross section, and calculate a similarity between the first image and a second image of a determined lumen intima contour; if the similarity is greater than or equal to a first preset value, starting the working process of the calculation module 200;

the calculating module 200 is configured to calculate an optimal torsion angle between the first image and the second image;

a first determining module 300, configured to determine an initial intraluminal contour of the first image according to the intraluminal contour of the second image and the optimal torsion angle;

a fitting module 400, configured to fit the initial lumen intima contour line based on the initial lumen intima contour line to obtain a final lumen intima contour line of the first image.

The blood vessel lumen intima contour extraction device provided by the embodiment of the application, when the similarity between the second image and the first image is larger than or equal to the first preset value, the initial lumen intima contour line of the first image is determined based on the lumen intima contour line of the second image, the clustering operation of pixel points in the first image is not needed again, the speed of determining the initial lumen intima contour line of the first image is improved, and the extraction speed of the blood vessel lumen intima contour is further improved. Otherwise, the initial lumen intima contour line of the first image is determined by other methods, so that the problem of inaccurate fitting caused by large image change is avoided, and the accuracy of the extraction of the lumen intima contour line of the blood vessel is improved. Meanwhile, when the similarity is larger than or equal to the first preset value, the optimal torsion angle between the second image and the first image is calculated, the lumen intima contour line of the second image and the optimal torsion angle are used for determining the initial lumen intima contour line of the first image, the problem of IVUS image torsion caused by heart beating is solved, and the accuracy of blood vessel lumen intima contour extraction is further improved. Therefore, the blood vessel lumen intima contour extraction device provided by the embodiment of the application determines the initial lumen intima contour line of the first image by using the lumen intima contour line of the second image and the optimal torsion angle when the similarity between the second image and the first image is greater than or equal to the first preset value, simplifies the image processing flow and improves the speed and accuracy of blood vessel lumen intima contour extraction.

On the basis of the foregoing embodiment, as a preferred implementation, the obtaining module 100 includes:

an acquisition unit configured to acquire a first image;

the first determining unit is used for determining a coordinate range corresponding to the region of interest based on the lumen intima contour line of the second image;

an extraction unit, configured to extract a first region of interest in the second image according to the coordinate range, and extract a second region of interest in the first image according to the coordinate range;

a first calculation unit configured to calculate a similarity between the first region of interest and the second region of interest as a similarity between the first image and the second image.

On the basis of the above embodiment, as a preferred implementation, the first determining unit includes:

the calculating subunit is used for calculating the maximum distance between the contour point in the lumen intima contour line of the second image and the image center of the second image, and taking the sum of the maximum distance and a preset offset distance as the target side length;

the determining subunit is used for determining a target square area and taking the coordinate range of the target square area as the coordinate range corresponding to the interested area; and the center of the target square area is the image center, and the side length is the target side length.

On the basis of the above embodiment, as a preferred implementation, the computing module 200 includes:

a second determination unit configured to determine an initial torsion angle range and an initial angle interval, and determine a target angle based on the initial angle interval within the initial torsion angle range;

a second calculation unit, configured to calculate a similarity between the first image and the second image at each of the target angles, and determine a maximum similarity among all the similarities;

the judging unit is used for judging whether preset conditions are met or not; the preset condition comprises that the maximum similarity is greater than or equal to a second preset value, or the updating times are greater than or equal to preset times; if yes, starting the working process of the third determination unit; if not, starting the working process of the updating unit;

a third determining unit, configured to use the target angle corresponding to the maximum similarity as an optimal torsion angle of the first image and the second image;

the updating unit is used for updating the initial torsion angle range and the initial angle interval, increasing the updating times by one, re-determining the target angle based on the updated angle interval in the updated torsion angle range, and restarting the working process of the second calculating unit; the updated torsion angle range is smaller than the torsion angle range before updating, the updated angle interval is smaller than the angle interval before updating, and the number of the target angles before updating is the same as that of the target angles after updating.

On the basis of the foregoing embodiment, as a preferred implementation manner, the first determining module 300 is specifically a module that calculates coordinates of contour points in the first image by using a contour point coordinate calculation formula, and determines a connection line of the contour points as an initial lumen intima contour line;

the contour point coordinate calculation formula specifically includes:

On the basis of the above embodiment, as a preferred implementation, the method further includes:

a second determining module, configured to determine a lumen wall area in the first image and determine an initial lumen intima contour line corresponding to the lumen wall area if the similarity is smaller than a first preset value; and fitting the initial lumen intima contour line to obtain a final lumen intima contour line of the first image.

On the basis of the foregoing embodiment, as a preferred implementation manner, the fitting module 400 is specifically a module that fits the initial intraluminal contour line by using a vector field convolution active contour model to obtain a final intraluminal contour line of the first image.

With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.

Based on the hardware implementation of the program module, and in order to implement the method of the embodiment of the present application, an embodiment of the present application further provides an ultrasound apparatus, and fig. 12 is a structural diagram of an ultrasound apparatus according to an exemplary embodiment, as shown in fig. 12, the ultrasound apparatus includes:

a communication interface 1 capable of information interaction with other devices such as network devices and the like;

and the processor 2 is connected with the communication interface 1 to realize information interaction with other equipment, and is used for executing the blood vessel lumen intima contour extraction method provided by one or more technical schemes when running a computer program. And the computer program is stored on the memory 3;

and the display 4 is used for displaying the intravascular ultrasonic image and the lumen intima contour line in the intravascular ultrasonic image.

Of course, in practice, the various components of the ultrasound device are coupled together by a bus system 5. It will be appreciated that the bus system 5 is used to enable connection communication between these components. The bus system 5 comprises, in addition to a data bus, a power bus, a control bus and a status signal bus. For clarity of illustration, however, the various buses are labeled as bus system 5 in fig. 12.

The memory 3 in the embodiment of the present application is used to store various types of data to support the operation of the ultrasound apparatus. Examples of such data include: any computer program for operating on an ultrasound device.

It will be appreciated that the memory 3 may be either volatile memory or nonvolatile memory, and may include both volatile and nonvolatile memory. Among them, the nonvolatile Memory may be a Read Only Memory (ROM), a Programmable Read Only Memory (PROM), an Erasable Programmable Read-Only Memory (EPROM), an Electrically Erasable Programmable Read-Only Memory (EEPROM), a magnetic random access Memory (FRAM), a Flash Memory (Flash Memory), a magnetic surface Memory, an optical disk, or a Compact Disc Read-Only Memory (CD-ROM); the magnetic surface storage may be disk storage or tape storage. Volatile Memory can be Random Access Memory (RAM), which acts as external cache Memory. By way of illustration and not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), Enhanced Synchronous Dynamic Random Access Memory (Enhanced DRAM), Synchronous Dynamic Random Access Memory (SLDRAM), Direct Memory (DRmb Access), and Random Access Memory (DRAM). The memory 2 described in the embodiments of the present application is intended to comprise, without being limited to, these and any other suitable types of memory.

The method disclosed in the above embodiment of the present application may be applied to the processor 2, or implemented by the processor 2. The processor 2 may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware or instructions in the form of software in the processor 2. The processor 2 described above may be a general purpose processor, a DSP, or other programmable logic device, discrete gate or transistor logic device, discrete hardware components, or the like. The processor 2 may implement or perform the methods, steps and logic blocks disclosed in the embodiments of the present application. A general purpose processor may be a microprocessor or any conventional processor or the like. The steps of the method disclosed in the embodiments of the present application may be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software modules may be located in a storage medium located in the memory 3, and the processor 2 reads the program in the memory 3 and in combination with its hardware performs the steps of the aforementioned method.

When the processor 2 executes the program, the corresponding processes in the methods according to the embodiments of the present application are realized, and for brevity, are not described herein again.

In an exemplary embodiment, the present application further provides a storage medium, i.e. a computer storage medium, specifically a computer readable storage medium, for example, including a memory 3 storing a computer program, which can be executed by a processor 2 to implement the steps of the foregoing method. The computer readable storage medium may be Memory such as FRAM, ROM, PROM, EPROM, EEPROM, Flash Memory, magnetic surface Memory, optical disk, or CD-ROM.

Those of ordinary skill in the art will understand that: all or part of the steps for implementing the method embodiments may be implemented by hardware related to program instructions, and the program may be stored in a computer readable storage medium, and when executed, the program performs the steps including the method embodiments; and the aforementioned storage medium includes: a removable storage device, a ROM, a RAM, a magnetic or optical disk, or various other media that can store program code.

Alternatively, the integrated units described above in the present application may be stored in a computer-readable storage medium if they are implemented in the form of software functional modules and sold or used as independent products. Based on such understanding, the technical solutions of the embodiments of the present application may be essentially or partially implemented in the form of a software product, which is stored in a storage medium and includes several instructions to enable an ultrasound device (which may be a personal computer, a server, or a network device) to execute all or part of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: a removable storage device, a ROM, a RAM, a magnetic or optical disk, or various other media that can store program code.

The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A method for extracting a contour of an intima of a blood vessel lumen is characterized by comprising the following steps:

2. The method for extracting the intima contour of a blood vessel according to claim 1, wherein calculating the similarity between the first image and the second image of the determined intima contour comprises:

3. The method for extracting the contour of the intima of a blood vessel according to claim 2, wherein the step of determining the coordinate range corresponding to the region of interest based on the contour of the intima of the lumen of the second image comprises the following steps:

4. The method for extracting the contour of the endovascular lumen according to claim 1, wherein calculating the optimal torsion angle of the first image and the second image comprises:

5. The method for extracting the contour of the intima of a blood vessel according to claim 1, wherein the step of determining the initial contour of the intima of the lumen of the first image according to the contour of the intima of the lumen of the second image and the optimal torsion angle comprises the following steps:

6. The method for extracting the contour of the endovascular lumen according to claim 1, further comprising:

if the similarity is smaller than a first preset value, determining a lumen wall area in the first image, and determining an initial lumen intima contour line corresponding to the lumen wall area.

7. The method for extracting the contour of the intima of a blood vessel according to any one of claims 1 to 6, wherein the step of fitting the initial intima contour line based on the initial intima contour line to obtain a final intima contour line of the lumen of the first image comprises the steps of:

8. A blood vessel lumen intimal contour extraction device, comprising:

9. An ultrasound device, comprising:

a memory for storing a computer program;

a processor for implementing the steps of the vessel lumen intima contour extraction method as claimed in any one of claims 1 to 7 when the computer program is executed;

10. A computer-readable storage medium, having stored thereon a computer program which, when being executed by a processor, carries out the steps of the vessel lumen intima contour extraction method as defined in any one of claims 1 to 7.