CN117115186B - Cardiovascular segmentation method based on region growth - Google Patents
Cardiovascular segmentation method based on region growth Download PDFInfo
- Publication number
- CN117115186B CN117115186B CN202311384642.1A CN202311384642A CN117115186B CN 117115186 B CN117115186 B CN 117115186B CN 202311384642 A CN202311384642 A CN 202311384642A CN 117115186 B CN117115186 B CN 117115186B
- Authority
- CN
- China
- Prior art keywords
- target
- preset
- initial seed
- seed point
- searching
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000002526 effect on cardiovascular system Effects 0.000 title claims abstract description 69
- 238000000034 method Methods 0.000 title claims abstract description 36
- 230000011218 segmentation Effects 0.000 title claims abstract description 26
- 230000002792 vascular Effects 0.000 claims abstract description 18
- 238000012216 screening Methods 0.000 claims abstract description 13
- 238000012545 processing Methods 0.000 claims abstract description 8
- 238000004458 analytical method Methods 0.000 claims description 6
- 210000004204 blood vessel Anatomy 0.000 description 9
- 230000000694 effects Effects 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 2
- 238000005481 NMR spectroscopy Methods 0.000 description 1
- 241001632422 Radiola linoides Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000002604 ultrasonography Methods 0.000 description 1
- 208000019553 vascular disease Diseases 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of image region segmentation, in particular to a cardiovascular segmentation method based on region growth, which comprises the following steps: grouping the pixel points in the acquired target cardiovascular image; determining an initial seed point; screening out preset extending directions different from all preset searching directions; respectively analyzing and processing gray level change rules of each target extending direction and each preset searching direction of each initial seed point; determining the gray level change consistency degree of each initial seed point between each target extending direction and each preset searching direction; determining a target area growth threshold value of each initial seed point in each preset searching direction; and carrying out region growth on the target cardiovascular image according to all the initial seed points and the target region growth threshold values of the initial seed points in all the preset search directions to obtain a target vascular region. According to the invention, the accuracy of cardiovascular segmentation is improved by carrying out region segmentation on the target cardiovascular image.
Description
Technical Field
The invention relates to the technical field of image region segmentation, in particular to a cardiovascular segmentation method based on region growth.
Background
Cardiovascular consists of heart and blood vessels, and in order for a physician to analyze vascular disease, it is often necessary to segment the vascular region from the cardiovascular image. Currently, when an image is segmented, the following methods are generally adopted: and carrying out region growth on the image by presetting a region growth threshold value to obtain a required region.
However, when region growing is performed on a cardiovascular image by presetting a region growing threshold, there are often the following technical problems:
because the preset region growing threshold is often a region growing threshold preset according to manual experience, the setting of the preset region growing threshold is often affected by artificial subjective factors, and the accuracy of the setting of the preset region growing threshold is often lower, so that the accuracy of cardiovascular segmentation is lower.
Disclosure of Invention
The summary of the invention is provided to introduce a selection of concepts in a simplified form that are further described below in the detailed description. The summary of the invention is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.
In order to solve the technical problem of lower accuracy of cardiovascular segmentation, the invention provides a cardiovascular segmentation method based on region growth.
The invention provides a cardiovascular segmentation method based on region growth, which comprises the following steps:
acquiring a target cardiovascular image, and grouping pixel points in the target cardiovascular image to obtain a pixel point group;
determining a pixel point with the minimum gray value in each pixel point group as an initial seed point;
screening preset extending directions different from all preset searching directions from a preset extending direction set to be used as target extending directions;
gray level change rule analysis processing is carried out on each target extending direction and each preset searching direction of each initial seed point respectively, so that gray level change characteristics of each initial seed point in each target extending direction and each preset searching direction are obtained;
determining the gray level change consistency degree of each initial seed point between the target extending direction and the preset searching direction according to the gray level change characteristics of each initial seed point in each target extending direction and each preset searching direction and the included angle between the target extending direction and the preset searching direction;
Correcting a preset region growth threshold according to the gray level change consistency degree of each initial seed point in all target extending directions and each preset searching direction to obtain a target region growth threshold of the initial seed point in the preset searching direction;
and carrying out region growth on the target cardiovascular image according to all initial seed points and target region growth thresholds of the initial seed points in all preset searching directions to obtain a target vascular region.
Optionally, the gray level change rule analysis processing is performed on each target extending direction and each preset searching direction of each initial seed point to obtain gray level change characteristics of each initial seed point in each target extending direction and each preset searching direction, including:
screening a first preset number of pixel points closest to the initial seed point in the target extending direction of the initial seed point to serve as extending pixel points, and obtaining an extending pixel point set in the target extending direction of the initial seed point;
screening out a second preset number of pixels closest to the initial seed point from the preset searching direction of the initial seed point to serve as searching pixels, and obtaining a searching pixel set in the preset searching direction of the initial seed point;
Determining gray scale change characteristics of the initial seed point in the target extending direction according to gray scale values corresponding to all extending pixel points in the extending pixel point set;
and determining the gray level change characteristics of the initial seed point in the preset searching direction according to gray level values corresponding to all the searching pixel points in the searching pixel point set.
Optionally, the determining, according to gray values corresponding to all extended pixels in the extended pixel set, gray change characteristics of the initial seed point in the target extension direction includes:
determining the average value of the absolute values of the difference values of the gray values corresponding to the initial seed points and the gray values corresponding to all the extension pixel points in the extension pixel point set as a first change index of the initial seed points in the target extension direction;
and determining the gray scale change characteristics of the initial seed point in the target extending direction according to the gray scale values corresponding to the first change index and all the extending pixel points in the extending pixel point set.
Optionally, the formula corresponding to the gray scale variation characteristic of the initial seed point in the target extending direction is:
；/>The method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Is the gray level change characteristic of the ith initial seed point in the jth target extending direction; i is the sequence number of the initial seed point; j is the sequence number of the extending direction of the target; />Is the first change index of the ith initial seed point in the jth target extending direction; />Is the number of extension pixel points in the extension pixel point set of the ith initial seed point in the jth target extension direction; />Is the gray value corresponding to the ith initial seed point；/>The gray value corresponding to the a-th extending pixel point in the extending pixel point set of the i-th initial seed point in the j-th target extending direction; />The gray value corresponding to the (a+1) th extended pixel point in the extension pixel point set of the (i) th initial seed point in the (j) th target extension direction; a is the serial number of the extending pixel point in the extending pixel point set of the ith initial seed point in the jth target extending direction; />Taking an absolute value function; />Is a preset factor greater than 0.
Optionally, the determining, according to gray values corresponding to all search pixels in the set of search pixels, gray change characteristics of the initial seed point in the preset search direction includes:
Determining the average value of the difference absolute values of the gray values corresponding to the initial seed points and the gray values corresponding to all the searching pixel points in the searching pixel point set as a second change index of the initial seed points in the preset searching direction;
and determining the gray scale change characteristics of the initial seed point in the preset searching direction according to the gray scale values corresponding to the second change index and all the searching pixel points in the searching pixel point set.
Optionally, the formula corresponding to the gray scale variation feature of the initial seed point in the preset search direction is:
；/>the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Is the gray level change characteristic of the ith initial seed point in the b preset searching direction; i is the sequence number of the initial seed point; b is a serial number of a preset searching direction; />Is a second variation index of the ith initial seed point in the b preset searching direction; />The number of the searching pixel points in the searching pixel point set of the ith initial seed point in the b preset searching direction; />Is the gray value corresponding to the ith initial seed point; />The gray value corresponding to the c searching pixel point in the searching pixel point set of the ith initial seed point in the b preset searching direction; / >The gray value corresponding to the (c+1) th searching pixel point in the searching pixel point set of the (i) th initial seed point in the (b) th preset searching direction; c is the serial number of the searching pixel point in the searching pixel point set of the ith initial seed point in the b preset searching direction; />Taking an absolute value function; />Is a preset factor greater than 0.
Optionally, the determining, according to the gray level change characteristics of each initial seed point in each target extending direction and each preset searching direction and the included angle between the target extending direction and the preset searching direction, the gray level change consistency degree of the initial seed point between the target extending direction and the preset searching direction includes:
determining the absolute value of the difference value of the gray level change characteristics of the initial seed point in the target extending direction and the preset searching direction as a characteristic difference index between the initial seed point in the target extending direction and the preset searching direction;
determining an included angle between the target extending direction and the preset searching direction as a target included angle;
and determining the gray level change consistency degree of the initial seed point between the target extending direction and the preset searching direction according to the characteristic difference index and the target included angle, wherein the characteristic difference index and the target included angle are in negative correlation with the gray level change consistency degree.
Optionally, the correcting the growth threshold of the preset area according to the gray level variation consistency degree of each initial seed point between all the target extending directions and each preset searching direction to obtain the growth threshold of the target area of the initial seed point in the preset searching direction includes:
screening out the maximum gray level change consistency degree from the gray level change consistency degrees of the initial seed points in all target extending directions and the preset searching directions, and taking the maximum gray level change consistency degree as the target consistency degree of the initial seed points in the preset searching directions;
determining the product of the target consistency degree and the preset region growth threshold value as the threshold value variation of the initial seed point in the preset searching direction;
and determining the sum of the preset region growing threshold and the threshold variation as a target region growing threshold of the initial seed point in the preset searching direction.
Optionally, the performing region growing on the target cardiovascular image according to all initial seed points and target region growing thresholds thereof in all preset searching directions to obtain a target vascular region includes:
Obtaining new seed points in the process of carrying out region growth on the target cardiovascular image according to all initial seed points and target region growth thresholds of the initial seed points in all preset searching directions;
determining the gray level change consistency degree of each new seed point between each target extending direction and each preset searching direction;
determining a target area growth threshold value of each new seed point in the preset search direction according to the gray level change consistency degree of each new seed point in all target extension directions and each preset search direction;
and carrying out region growth on the target cardiovascular image according to all the seed points and the target region growth threshold values of the seed points in all the preset search directions to obtain a target vascular region.
Optionally, a formula corresponding to the target area growth threshold of the new seed point in the preset search direction is:
the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Is the target area growth threshold of the xth new seed point in the b preset searching direction; />Is the maximum value of the gray level variation consistency degree of the xth new seed point between all the target extending directions and the b preset searching direction; />Growing the xth new seed point as a target area growth threshold of the seed point in the b preset searching direction; x is the sequence number of the new seed point; b is a serial number of a preset search direction.
The invention has the following beneficial effects:
according to the cardiovascular segmentation method based on region growth, the region segmentation is carried out on the target cardiovascular image, so that the cardiovascular segmentation is realized, the technical problem of low accuracy of the cardiovascular segmentation is solved, and the accuracy of the cardiovascular segmentation is improved. First, the pixels in the target cardiovascular image are grouped, so that initial seed points can be conveniently screened from each pixel group later. Then, because the gray value corresponding to the blood vessel is relatively low, the pixel point with the smallest gray value in the pixel point group is often the pixel point in the blood vessel, and the pixel point is taken as an initial seed point, so that the growth of a subsequent region can be facilitated, and a target blood vessel region is obtained. Continuing, since the gray level change rule of the initial seed points in each target extending direction and each preset searching direction often affects the setting of the growth threshold of the subsequent region, the gray level change feature of each initial seed point in each target extending direction and each preset searching direction is quantified, so that the target region growth threshold can be accurately set for each initial seed point in each preset searching direction. Then, comprehensively considering the gray level change characteristics of each initial seed point in each target extending direction and each preset searching direction and the included angle between the target extending direction and the preset searching direction, the gray level change consistency degree of the initial seed point in the target extending direction and the preset searching direction can be quantified, and the larger the value is, the larger the area growth threshold value can be set for the initial seed point in the preset searching direction. Therefore, the gray level change consistency degree of each initial seed point in all the target extending directions and each preset searching direction is comprehensively considered, and the preset area growth threshold value is corrected, so that the target area growth threshold value of the initial seed point in the preset searching direction can be obtained. Finally, based on all initial seed points and target region growth thresholds thereof in all preset search directions, region growth is carried out on the target cardiovascular image, so that a target vascular region can be obtained, and compared with the region growth of the cardiovascular image through the preset region growth thresholds, the region growth method and the region growth device provided by the invention quantify indexes set by a plurality of influence region growth thresholds, such as gray level change consistency degree, relatively objectively quantify the target region growth thresholds in each preset search direction for each initial seed point, and reduce the influence of artificial subjective factors to a certain extent, thereby improving the accuracy of region growth on the target cardiovascular image, and further improving the accuracy of determination of the target vascular region.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a region-growth based cardiovascular segmentation method of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the present invention to achieve the preset purpose, the following detailed description is given below of the specific implementation, structure, features and effects of the technical solution according to the present invention with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
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.
The invention provides a cardiovascular segmentation method based on region growth, which comprises the following steps:
acquiring a target cardiovascular image, and grouping pixel points in the target cardiovascular image to obtain a pixel point group;
determining a pixel point with the minimum gray value in each pixel point group as an initial seed point;
screening preset extending directions different from all preset searching directions from a preset extending direction set to be used as target extending directions;
gray level change rule analysis processing is carried out on each target extending direction and each preset searching direction of each initial seed point respectively, so that gray level change characteristics of each initial seed point in each target extending direction and each preset searching direction are obtained;
determining the gray level change consistency degree of the initial seed points between the target extending direction and the preset searching direction according to the gray level change characteristics of each initial seed point in each target extending direction and each preset searching direction and the included angle between the target extending direction and the preset searching direction;
correcting the preset region growing threshold according to the gray level change consistency degree of each initial seed point in all the target extending directions and each preset searching direction to obtain the target region growing threshold of the initial seed point in the preset searching direction;
And carrying out region growth on the target cardiovascular image according to all the initial seed points and the target region growth threshold values of the initial seed points in all the preset search directions to obtain a target vascular region.
The following detailed development of each step is performed:
referring to fig. 1, a flow of some embodiments of a region-growth based cardiovascular segmentation method according to the present invention is shown. The cardiovascular segmentation method based on region growth comprises the following steps:
step S1, acquiring a target cardiovascular image, and grouping pixel points in the target cardiovascular image to obtain a pixel point group.
In some embodiments, a target cardiovascular image may be acquired, and pixels in the target cardiovascular image may be grouped to obtain a pixel group.
The target cardiovascular image may be a cardiovascular image after image preprocessing. Image preprocessing may include, but is not limited to: graying, image enhancement and denoising. The pixel point group may include: a row of pixels in the target cardiovascular image.
It should be noted that, grouping the pixels in the target cardiovascular image may facilitate subsequent screening of the initial seed points from each pixel group.
As an example, this step may include the steps of:
first, a cardiovascular image is acquired as an initial cardiovascular image by an image acquisition device.
Wherein the image acquisition device may be a device for image acquisition. For example, the image acquisition device may be, but is not limited to: an X-ray imager, an ultrasound imager or a nuclear magnetic resonance imager.
And secondly, graying the initial cardiovascular image, and taking the image obtained after graying as a target cardiovascular image.
Thirdly, each row of pixel points in the target cardiovascular image form a pixel point group.
And S2, determining the pixel point with the minimum gray value in each pixel point group as an initial seed point.
In some embodiments, the pixel point with the smallest gray value in each pixel point group may be determined as the initial seed point.
It should be noted that, because the gray value corresponding to the blood vessel is relatively low, the pixel point with the smallest gray value in the pixel point group is often the pixel point in the blood vessel, and the pixel point is used as the initial seed point, so that the growth of the subsequent region can be facilitated, and the target blood vessel region can be obtained.
As an example, a pixel having the smallest gray value may be selected from each pixel group as an initial seed point.
Step S3, screening out preset extending directions different from all preset searching directions from the preset extending direction set, and taking the preset extending directions as target extending directions.
In some embodiments, a preset extending direction different from all preset searching directions may be selected from the preset extending direction set as the target extending direction.
The preset extending direction may be a preset direction for comparison with a preset searching direction. The preset search direction may be preset, that is, a search direction when region growing is performed, that is, a growth direction when region growing is performed. The angle between adjacent preset extension directions may be smaller than the angle between adjacent preset search directions.
As an example, if the neighborhood when the region growing is performed is a 4 neighborhood, the included angle between adjacent preset search directions may be 90 °, and the total of 4 preset search directions may be: 0 ° direction, 90 ° direction, 180 ° direction, and 270 ° direction. Because the included angle between the adjacent preset extending directions can be smaller than the included angle between the adjacent preset searching directions, each preset extending direction included in the preset extending direction set can be in sequence: 0 ° direction, 45 ° direction, 90 ° direction, 135 ° direction, 180 ° direction, 225 ° direction, 270 ° direction, and 315 ° direction. In this case, the total of 4 target extending directions are 45 °, 135 °, 225 °, and 315 °.
And S4, respectively analyzing and processing the gray level change rule of each initial seed point in each target extending direction and each preset searching direction to obtain gray level change characteristics of each initial seed point in each target extending direction and each preset searching direction.
In some embodiments, gray level change rule analysis processing may be performed on each target extending direction and each preset searching direction of each initial seed point, so as to obtain gray level change characteristics of each initial seed point in each target extending direction and each preset searching direction.
It should be noted that, since the gray level change rule of the initial seed point in each target extending direction and each preset searching direction often affects the setting of the growth threshold of the subsequent area, the gray level change feature of each initial seed point in each target extending direction and each preset searching direction is quantified, so that the target area growth threshold can be accurately set for each initial seed point in each preset searching direction.
As an example, this step may include the steps of:
the first step is to screen out a first preset number of pixels closest to the initial seed point in the target extending direction of the initial seed point as extending pixels, and obtain an extending pixel point set in the target extending direction of the initial seed point.
The first preset number may be a preset number. For example, the first preset number may be 20.
And a second step of screening out a second preset number of pixels closest to the initial seed point in the preset searching direction of the initial seed point as searching pixels to obtain a searching pixel point set in the preset searching direction of the initial seed point.
Wherein the second preset number may be a preset number. The first preset number may be equal to the second preset number. For example, the second preset number may be 20.
The third step of determining the gray scale variation characteristic of the initial seed point in the target extending direction according to the gray scale values corresponding to all the extending pixel points in the extending pixel point set may include the following substeps:
and a first sub-step of determining the average value of the absolute value of the difference between the gray value corresponding to the initial seed point and the gray value corresponding to all the extension pixel points in the extension pixel point set as a first change index of the initial seed point in the target extension direction.
A second substep, determining, according to the gray values corresponding to the first change index and all the extended pixels in the extended pixel set, a formula corresponding to the gray change feature of the initial seed point in the target extension direction may be:
；/>The method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Is the ith initial seed point in the jth target extension directionGray scale variation characteristics. i is the sequence number of the initial seed point. j is the sequence number of the target extension direction. />Is the first index of change of the ith initial seed point in the jth target extension direction. />Is the number of extended pixels in the set of extended pixels for the ith initial seed point in the jth target extension direction. />Is the gray value corresponding to the i-th initial seed point. />Is the gray value corresponding to the a-th extending pixel point in the extending pixel point set of the i-th initial seed point in the j-th target extending direction. />Is the gray value corresponding to the (a+1) th extended pixel point in the extended pixel point set of the (i) th initial seed point in the (j) th target extending direction. a is the serial number of the extension pixel point in the extension pixel point set of the ith initial seed point in the jth target extension direction. />Is a function of absolute value. />Is a factor greater than 0 preset, mainly for preventing denominator from being 0, such as ++>May be 0.01.
When the following is performedThe larger the gray scale change of the ith initial seed point in the jth target extending direction is, the larger the gray scale change tends to be The more dissimilar the gray value between the i-th initial seed point and the pixel point in the j-th target extension direction is explained. When->The larger the gradation difference between the a-th extended pixel point and the a+1-th extended pixel point is, the larger the gradation change between the a-th extended pixel point and the a+1-th extended pixel point is, and the gradation value of the extended pixel point is gradually changed. Thus (S)>The gray level change characteristic of the ith initial seed point in the jth target extending direction can be characterized, and the larger the value of the gray level change characteristic is, the larger the gray level change degree of the ith initial seed point in the jth target extending direction is usually indicated.
Fourth, determining the gray scale variation characteristics of the initial seed point in the preset searching direction according to the gray scale values corresponding to all the searching pixels in the searching pixel point set may include the following sub-steps:
and a first sub-step of determining the average value of the absolute value of the difference between the gray value corresponding to the initial seed point and the gray value corresponding to all the searching pixels in the searching pixel point set as a second change index of the initial seed point in the preset searching direction.
A second substep, determining, according to the gray values corresponding to the second change index and all the search pixels in the set of search pixels, a formula corresponding to the gray change feature of the initial seed point in the preset search direction may be:
；/>the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Is the ith initial seed point at the b < th > pre-determinedLet the gray scale variation feature in the search direction. i is the sequence number of the initial seed point. b is a serial number of a preset search direction. />Is a second variation index of the ith initial seed point in the b-th preset search direction. />Is the number of search pixels in the search pixel set of the ith initial seed point in the b-th preset search direction. />Is the gray value corresponding to the i-th initial seed point. />The gray value corresponding to the c-th searching pixel point in the searching pixel point set of the i-th initial seed point in the b-th preset searching direction. />The gray value corresponding to the (c+1) th searching pixel point in the searching pixel point set of the (i) th initial seed point in the (b) th preset searching direction. c is the serial number of the searching pixel point in the searching pixel point set of the ith initial seed point in the b preset searching direction. />Is a function of absolute value. / >Is a factor greater than 0 preset, mainly for preventing denominator from being 0, such as ++>May be 0.01.
When the following is performedThe larger the size, the more often the gray of the ith initial seed point in the b-th preset search direction is describedThe larger the degree variation, the more dissimilar the gray value between the ith initial seed point and the pixel point in its b-th preset search direction tends to be. When->The larger the gradation difference between the c-th search pixel and the c+1th search pixel is, the larger the gradation change between the c-th search pixel and the c+1th search pixel is, and the gradation value of the search pixel is gradually changed. Thus (S)>The gray level change characteristic of the ith initial seed point in the b preset searching direction can be represented, and the larger the value of the gray level change characteristic is, the larger the gray level change degree of the ith initial seed point in the b preset searching direction is often explained.
And S5, determining the gray level change consistency degree of the initial seed points between the target extending direction and the preset searching direction according to the gray level change characteristics of each initial seed point in each target extending direction and each preset searching direction and the included angle between the target extending direction and the preset searching direction.
In some embodiments, the gray level variation consistency degree of the initial seed point between the target extending direction and the preset searching direction may be determined according to the gray level variation characteristic of each initial seed point in each target extending direction and each preset searching direction and the included angle between the target extending direction and the preset searching direction.
It should be noted that, by comprehensively considering the gray level change characteristics of each initial seed point in each target extending direction and each preset searching direction and the included angle between the target extending direction and the preset searching direction, the gray level change consistency degree of the initial seed point between the target extending direction and the preset searching direction can be quantified, and the larger the value is, the larger the area growth threshold value can be set for the initial seed point in the preset searching direction.
As an example, this step may include the steps of:
the first step is to determine the absolute value of the difference value of the gray level change characteristic of the initial seed point in the target extending direction and the preset searching direction as the characteristic difference index between the target extending direction and the preset searching direction of the initial seed point.
And secondly, determining an included angle between the target extending direction and the preset searching direction as a target included angle.
And thirdly, determining the gray level change consistency degree of the initial seed point between the target extending direction and the preset searching direction according to the characteristic difference index and the target included angle.
Wherein, the characteristic difference index and the target included angle can be inversely related to the gray level variation consistency degree.
For example, the formula for determining the correspondence of the gray level variation coincidence degree of the initial seed point between the target extending direction and the preset searching direction may be:
the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Is the degree of coincidence of the gray level change of the i-th initial seed point between the j-th target extending direction and the b-th preset searching direction. />Is a normalization function. />Is a function of absolute value. />The included angle between the jth target extending direction and the b preset searching direction is the target included angle.Is the ith initial seed pointGray scale change characteristics in the j-th target extending direction. />Is the gray level change characteristic of the ith initial seed point in the b preset search direction. i is the sequence number of the initial seed point. j is the sequence number of the target extension direction. b is a serial number of a preset search direction. / >Is a characteristic difference index of the ith initial seed point between the jth target extending direction and the b preset searching direction.
When the following is performedThe smaller the time, the more similar the jth target extending direction and the b preset searching direction are often explained. When->The smaller the time, the more similar the gray scale change characteristics of the ith initial seed point between the jth target extending direction and the b preset searching direction are often explained. Thus->When the gray level change of the ith initial seed point between the jth target extending direction and the b preset searching direction is more similar, the jth target extending direction and the b preset searching direction are more similar, the gray level change of the ith initial seed point between the jth target extending direction and the b preset searching direction is more consistent, and the growth threshold value is moderately increased at the moment, so that the pixels with similar gray levels can be more easily divided into the same region by region growth, and the growth effect and clustering accuracy of the region growth can be improved. Thus->The larger it is often stated that the larger its corresponding region growth threshold needs to be.
And S6, correcting the preset region growing threshold according to the gray level change consistency degree of each initial seed point in all the target extending directions and each preset searching direction, and obtaining the target region growing threshold of the initial seed point in the preset searching direction.
In some embodiments, the preset region growing threshold may be corrected according to the gray level variation consistency degree of each initial seed point between all the target extending directions and each preset searching direction, so as to obtain the target region growing threshold of the initial seed point in the preset searching direction.
The preset region growing threshold may be a preset threshold for region growing, that is, a preset region growing threshold. For example, the preset region growing threshold may be a region growing threshold set according to manual experience.
It should be noted that, the gray level change consistency degree of each initial seed point in all the target extending directions and each preset searching direction is comprehensively considered, and the preset area growth threshold is corrected, so that the target area growth threshold with relatively accurate initial seed points in the preset searching directions can be obtained.
As an example, this step may include the steps of:
the first step is to screen out the maximum gray level change consistency degree from the gray level change consistency degree of the initial seed point between all the target extending directions and the preset searching direction, and the maximum gray level change consistency degree is used as the target consistency degree of the initial seed point in the preset searching direction.
And a second step of determining the product of the target coincidence degree and the preset region growth threshold value as the threshold value variation of the initial seed point in the preset searching direction.
And thirdly, determining the sum of the preset region growing threshold and the threshold variation as a target region growing threshold of the initial seed point in the preset searching direction.
For example, the formula corresponding to the target region growth threshold for determining the initial seed point in the preset search direction may be:
the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Is the target area growth threshold of the ith initial seed point in the b preset search direction. d is a preset region growth threshold. />Is the target coincidence degree of the ith initial seed point in the b-th preset search direction, that is, the maximum value of the gray level variation coincidence degree of the ith initial seed point between all the target extending directions and the b-th preset search direction. />Is the threshold variation of the ith initial seed point in the b-th preset search direction. i is the sequence number of the initial seed point. b is a serial number of a preset search direction.
When the following is performedThe larger the i-th initial seed point is, the more consistent the gray level change between each target extending direction and the b-th preset searching direction is, and the larger the corresponding region growing threshold is required to be adjusted. Thus- >The adjusted region growing threshold set by the ith initial seed point in the b-th preset search direction may be characterized.
And S7, carrying out region growth on the target cardiovascular image according to all initial seed points and target region growth thresholds of the initial seed points in all preset search directions to obtain a target vascular region.
In some embodiments, the target cardiovascular image may be subjected to region growing according to all initial seed points and target region growing thresholds thereof in all preset searching directions, so as to obtain a target vascular region.
Wherein the target vascular region may be a vascular region in a target cardiovascular image.
As an example, this step may include the steps of:
according to all initial seed points and target region growth thresholds in all preset searching directions, new seed points are obtained in the process of carrying out region growth on the target cardiovascular images.
Wherein the new seed point may be a seed point newly grown during the region growing process.
The second step, the method for determining the gray level variation consistency degree of each new seed point between each target extending direction and each preset searching direction can be as follows: and taking the new seed point as an initial seed point, and executing the step S4 and the step S5, wherein the obtained gray level change consistency degree is the gray level change consistency degree of the new seed point between the target extending direction and the preset searching direction.
Thirdly, determining a formula corresponding to a target area growth threshold value of each new seed point in the preset searching direction according to the gray level change consistency degree of each new seed point between all the target extending directions and each preset searching direction, wherein the formula can be as follows:
the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Is the target area growth threshold of the xth new seed point in the b preset search direction. />Is the maximum value of the x new seed point in the gray level variation consistency degree between all the target extending directions and the b preset searching direction. />The method is to grow the xth new seed point into the target area growth threshold value of the seed point in the b preset searching direction. x is the sequence number of the new seed point. b is a serial number of a preset search direction.
When the following is performedThe larger the new seed point is, the more consistent the gray level change of the xth new seed point between each target extending direction and the b preset searching direction is, and the more the corresponding region growing threshold is required to be increased. Thus->The adjusted region growing threshold set by the xth new seed point in the b-th preset search direction may be characterized. />
And step four, carrying out region growth on the target cardiovascular image according to all seed points and target region growth thresholds thereof in all preset searching directions to obtain a target vascular region.
In order to make the target blood vessel region clearer, the target blood vessel region may be subjected to denoising and image enhancement.
In summary, based on all initial seed points and target region growth thresholds thereof in all preset search directions, region growth is performed on a target cardiovascular image, so that a target vascular region can be obtained, and compared with region growth performed on the cardiovascular image through the preset region growth thresholds, the method and the device provided by the invention quantify indexes set by a plurality of influence region growth thresholds, such as gray level change consistency degree, relatively objectively quantify the target region growth thresholds in each preset search direction for each initial seed point, and reduce the influence of artificial subjective factors to a certain extent, thereby improving the accuracy of region growth on the target cardiovascular image, and further improving the accuracy of target vascular region determination.
The above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the invention and are intended to be included within the scope of the invention.
Claims (5)
1. A cardiovascular segmentation method based on region growing, comprising the steps of:
acquiring a target cardiovascular image, and grouping pixel points in the target cardiovascular image to obtain a pixel point group;
determining a pixel point with the minimum gray value in each pixel point group as an initial seed point;
screening out preset extending directions which are different from all preset searching directions from a preset extending direction set to be used as target extending directions, wherein the preset extending directions are preset and are used for comparing with the preset searching directions; the preset searching direction is preset, and the searching direction is the searching direction when the region grows;
gray level change rule analysis processing is carried out on each target extending direction and each preset searching direction of each initial seed point respectively, so that gray level change characteristics of each initial seed point in each target extending direction and each preset searching direction are obtained;
determining the gray level change consistency degree of each initial seed point between the target extending direction and the preset searching direction according to the gray level change characteristics of each initial seed point in each target extending direction and each preset searching direction and the included angle between the target extending direction and the preset searching direction;
Correcting a preset region growth threshold according to the gray level change consistency degree of each initial seed point in all target extending directions and each preset searching direction to obtain a target region growth threshold of the initial seed point in the preset searching direction;
according to all initial seed points and target region growth thresholds of the initial seed points in all preset search directions, performing region growth on the target cardiovascular image to obtain a target vascular region;
the gray level change rule analysis processing is performed on each target extending direction and each preset searching direction of each initial seed point to obtain gray level change characteristics of each initial seed point in each target extending direction and each preset searching direction, including:
screening a first preset number of pixel points closest to the initial seed point in the target extending direction of the initial seed point to serve as extending pixel points, and obtaining an extending pixel point set in the target extending direction of the initial seed point;
screening out a second preset number of pixels closest to the initial seed point from the preset searching direction of the initial seed point to serve as searching pixels, and obtaining a searching pixel set in the preset searching direction of the initial seed point;
Determining gray scale change characteristics of the initial seed point in the target extending direction according to gray scale values corresponding to all extending pixel points in the extending pixel point set;
determining gray level change characteristics of the initial seed points in the preset searching direction according to gray level values corresponding to all searching pixel points in the searching pixel point set;
the determining the gray scale change characteristic of the initial seed point in the target extending direction according to the gray scale values corresponding to all the extending pixel points in the extending pixel point set comprises the following steps:
determining the average value of the absolute values of the difference values of the gray values corresponding to the initial seed points and the gray values corresponding to all the extension pixel points in the extension pixel point set as a first change index of the initial seed points in the target extension direction;
according to the gray scale values corresponding to the first change index and all the extension pixel points in the extension pixel point set, determining the gray scale change characteristics of the initial seed point in the target extension direction;
the formula corresponding to the gray scale change characteristic of the initial seed point in the target extending direction is as follows:
；/>the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Is the gray level change characteristic of the ith initial seed point in the jth target extending direction; i is the sequence number of the initial seed point; j is the sequence number of the extending direction of the target; / >Is the first change index of the ith initial seed point in the jth target extending direction; />Is the number of extension pixel points in the extension pixel point set of the ith initial seed point in the jth target extension direction; />Is the gray value corresponding to the ith initial seed point; />The gray value corresponding to the a-th extending pixel point in the extending pixel point set of the i-th initial seed point in the j-th target extending direction; />The gray value corresponding to the (a+1) th extended pixel point in the extension pixel point set of the (i) th initial seed point in the (j) th target extension direction; a is the serial number of the extending pixel point in the extending pixel point set of the ith initial seed point in the jth target extending direction; />Is taken absoluteA value function; />Is a factor greater than 0 set in advance;
the determining the gray level change characteristic of the initial seed point in the preset searching direction according to the gray level values corresponding to all the searching pixel points in the searching pixel point set comprises the following steps:
determining the average value of the difference absolute values of the gray values corresponding to the initial seed points and the gray values corresponding to all the searching pixel points in the searching pixel point set as a second change index of the initial seed points in the preset searching direction;
Determining gray scale change characteristics of the initial seed point in the preset searching direction according to gray scale values corresponding to the second change indexes and all searching pixel points in the searching pixel point set;
the formula corresponding to the gray level change characteristic of the initial seed point in the preset search direction is as follows:
；/>the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Is the gray level change characteristic of the ith initial seed point in the b preset searching direction; i is the sequence number of the initial seed point; b is a serial number of a preset searching direction; />Is a second variation index of the ith initial seed point in the b preset searching direction; />Is the searching image in the searching pixel point set of the ith initial seed point in the b preset searching directionThe number of pixels; />Is the gray value corresponding to the ith initial seed point; />The gray value corresponding to the c searching pixel point in the searching pixel point set of the ith initial seed point in the b preset searching direction; />The gray value corresponding to the (c+1) th searching pixel point in the searching pixel point set of the (i) th initial seed point in the (b) th preset searching direction; c is the serial number of the searching pixel point in the searching pixel point set of the ith initial seed point in the b preset searching direction; / >Taking an absolute value function; />Is a preset factor greater than 0.
2. The region growing-based cardiovascular segmentation method according to claim 1, wherein the determining the gray level variation consistency degree of the initial seed points between the target extending direction and the preset searching direction according to the gray level variation characteristic of each initial seed point in each target extending direction and each preset searching direction and the included angle between the target extending direction and the preset searching direction comprises:
determining the absolute value of the difference value of the gray level change characteristics of the initial seed point in the target extending direction and the preset searching direction as a characteristic difference index between the initial seed point in the target extending direction and the preset searching direction;
determining an included angle between the target extending direction and the preset searching direction as a target included angle;
and determining the gray level change consistency degree of the initial seed point between the target extending direction and the preset searching direction according to the characteristic difference index and the target included angle, wherein the characteristic difference index and the target included angle are in negative correlation with the gray level change consistency degree.
3. The cardiovascular segmentation method based on region growing according to claim 1, wherein the correcting the preset region growing threshold according to the gray level variation consistency degree of each initial seed point between all the target extending directions and each preset searching direction to obtain the target region growing threshold of the initial seed point in the preset searching direction comprises:
screening out the maximum gray level change consistency degree from the gray level change consistency degrees of the initial seed points in all target extending directions and the preset searching directions, and taking the maximum gray level change consistency degree as the target consistency degree of the initial seed points in the preset searching directions;
determining the product of the target consistency degree and the preset region growth threshold value as the threshold value variation of the initial seed point in the preset searching direction;
and determining the sum of the preset region growing threshold and the threshold variation as a target region growing threshold of the initial seed point in the preset searching direction.
4. The region-growing-based cardiovascular segmentation method according to claim 1, wherein the performing region growing on the target cardiovascular image according to all initial seed points and target region growing thresholds thereof in all preset search directions to obtain a target vascular region comprises:
Obtaining new seed points in the process of carrying out region growth on the target cardiovascular image according to all initial seed points and target region growth thresholds of the initial seed points in all preset searching directions;
determining the gray level change consistency degree of each new seed point between each target extending direction and each preset searching direction;
determining a target area growth threshold value of each new seed point in the preset search direction according to the gray level change consistency degree of each new seed point in all target extension directions and each preset search direction;
and carrying out region growth on the target cardiovascular image according to all the seed points and the target region growth threshold values of the seed points in all the preset search directions to obtain a target vascular region.
5. The region growing-based cardiovascular segmentation method according to claim 4, wherein the formula corresponding to the target region growing threshold of the new seed point in the preset search direction is:
the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Is the target area growth threshold of the xth new seed point in the b preset searching direction; />Is the maximum value of the gray level variation consistency degree of the xth new seed point between all the target extending directions and the b preset searching direction; / >Growing the xth new seed point as a target area growth threshold of the seed point in the b preset searching direction; x is the sequence number of the new seed point; b is a serial number of a preset search direction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311384642.1A CN117115186B (en) | 2023-10-25 | 2023-10-25 | Cardiovascular segmentation method based on region growth |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311384642.1A CN117115186B (en) | 2023-10-25 | 2023-10-25 | Cardiovascular segmentation method based on region growth |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117115186A CN117115186A (en) | 2023-11-24 |
| CN117115186B true CN117115186B (en) | 2024-02-02 |
Family
ID=88796987
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311384642.1A Active CN117115186B (en) | 2023-10-25 | 2023-10-25 | Cardiovascular segmentation method based on region growth |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117115186B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104715483A (en) * | 2015-03-20 | 2015-06-17 | 中国科学院自动化研究所 | Automatic segmenting method for lesion tissue in lung CT image |
| CN107045721A (en) * | 2016-10-24 | 2017-08-15 | 东北大学 | One kind extracts pulmonary vascular method and device from chest CT image |
| CN111951277A (en) * | 2020-07-28 | 2020-11-17 | 杭州电子科技大学 | Coronary artery segmentation method based on CTA image |
| CN115375719A (en) * | 2022-08-18 | 2022-11-22 | 华科精准（北京）医疗科技有限公司 | Blood vessel image segmentation method and device |
| CN116843714A (en) * | 2023-07-10 | 2023-10-03 | 复旦大学 | CT image vascular reconstruction method adopting region growing method |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7711165B2 (en) * | 2005-07-28 | 2010-05-04 | Siemens Medical Solutions Usa, Inc. | System and method for coronary artery segmentation of cardiac CT volumes |
| CN108986119B (en) * | 2018-07-25 | 2020-07-28 | 京东方科技集团股份有限公司 | Image segmentation method and device, computer equipment and readable storage medium |
-
2023
- 2023-10-25 CN CN202311384642.1A patent/CN117115186B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104715483A (en) * | 2015-03-20 | 2015-06-17 | 中国科学院自动化研究所 | Automatic segmenting method for lesion tissue in lung CT image |
| CN107045721A (en) * | 2016-10-24 | 2017-08-15 | 东北大学 | One kind extracts pulmonary vascular method and device from chest CT image |
| CN111951277A (en) * | 2020-07-28 | 2020-11-17 | 杭州电子科技大学 | Coronary artery segmentation method based on CTA image |
| CN115375719A (en) * | 2022-08-18 | 2022-11-22 | 华科精准（北京）医疗科技有限公司 | Blood vessel image segmentation method and device |
| CN116843714A (en) * | 2023-07-10 | 2023-10-03 | 复旦大学 | CT image vascular reconstruction method adopting region growing method |
Non-Patent Citations (3)
| Title |
|---|
| 基于区域生长法的三维心血管模型的分割;谭启路 等;《北京生物医学工程》;第35卷(第5期);第469-475页 * |
| 多模态融合技术三维重建岩上静脉;吴世源 等;《中国临床解剖学杂志》;第41卷(第3期);第270-276页 * |
| 改进区域生长法及其在冠状动脉造影图中的应用;於猛 等;《厦门大学学报（自然科学版）》;第52卷(第1期);第38-42页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117115186A (en) | 2023-11-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109480780B (en) | Evaluation method and system of stroke early warning system | |
| CN111784721B (en) | Ultrasonic endoscopic image intelligent segmentation and quantification method and system based on deep learning | |
| CN102068281B (en) | Processing method for space-occupying lesion ultrasonic images | |
| CN116109663B (en) | Stomach CT image segmentation method based on multi-threshold segmentation | |
| CN108664976B (en) | Super-pixel-based fuzzy spectral clustering brain tumor image automatic segmentation method | |
| CN113361623B (en) | Medical image classification method combining lightweight CNN with transfer learning | |
| CN117764864B (en) | Nuclear magnetic resonance tumor visual detection method based on image denoising | |
| CN115797352B (en) | Tongue picture image processing system for traditional Chinese medicine health-care physique detection | |
| CN111047607A (en) | Method for automatically segmenting coronary artery | |
| CN112529918B (en) | Method, device and equipment for segmenting brain room area in brain CT image | |
| CN109949288A (en) | Tumor type determines system, method and storage medium | |
| CN117689574B (en) | Medical image processing method for tumor radio frequency ablation diagnosis and treatment positioning | |
| CN116721099B (en) | Image segmentation method of liver CT image based on clustering | |
| CN117523235B (en) | A patient wound intelligent identification system for surgical nursing | |
| CN118172380B (en) | Orthopedics leg bone intelligent recognition segmentation method based on local threshold | |
| CN117115186B (en) | Cardiovascular segmentation method based on region growth | |
| CN110619633B (en) | Liver image segmentation method based on multipath filtering strategy | |
| Kalhor et al. | Assessment of histogram-based medical image contrast enhancement techniques; an implementation | |
| CN116485819A (en) | Ear-nose-throat examination image segmentation method and system | |
| CN113763409B (en) | Gaussian filtering-based multi-threshold segmentation method for high-dimensional brain nuclear magnetic image | |
| CN110647870B (en) | Method for calculating approximate entropy of resting state fMRI data based on sliding window | |
| CN113034451A (en) | Chest DR image identification method based on deep learning | |
| CN118397023B (en) | Trigeminal region segmentation method based on MRI | |
| CN117974692B (en) | Ophthalmic medical image processing method based on region growing | |
| CN117689552B (en) | Coronary angiography enhancement method for intracardiac interventional therapy |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |