CN116630358B - Threshold segmentation method for brain tumor CT image - Google Patents
Threshold segmentation method for brain tumor CT image Download PDFInfo
- Publication number
- CN116630358B CN116630358B CN202310912887.0A CN202310912887A CN116630358B CN 116630358 B CN116630358 B CN 116630358B CN 202310912887 A CN202310912887 A CN 202310912887A CN 116630358 B CN116630358 B CN 116630358B
- Authority
- CN
- China
- Prior art keywords
- value
- current
- representing
- pixel points
- connected region
- 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
- 238000000034 method Methods 0.000 title claims abstract description 62
- 230000011218 segmentation Effects 0.000 title claims abstract description 45
- 208000003174 Brain Neoplasms Diseases 0.000 title claims abstract description 22
- 206010028980 Neoplasm Diseases 0.000 claims abstract description 89
- 238000004891 communication Methods 0.000 claims abstract description 75
- 230000000877 morphologic effect Effects 0.000 claims abstract description 12
- 238000012545 processing Methods 0.000 claims abstract description 12
- 230000000694 effects Effects 0.000 claims abstract description 11
- 230000008569 process Effects 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000003708 edge detection Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 claims description 3
- 238000007917 intracranial administration Methods 0.000 abstract description 3
- 238000011156 evaluation Methods 0.000 abstract description 2
- 210000004556 brain Anatomy 0.000 description 7
- 238000013210 evaluation model Methods 0.000 description 7
- 230000008859 change Effects 0.000 description 5
- 210000004907 gland Anatomy 0.000 description 4
- 201000011510 cancer Diseases 0.000 description 3
- 238000013507 mapping Methods 0.000 description 3
- 208000012868 Overgrowth Diseases 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 210000004204 blood vessel Anatomy 0.000 description 2
- 230000000762 glandular Effects 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 210000004881 tumor cell Anatomy 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000005013 brain tissue Anatomy 0.000 description 1
- 210000004027 cell Anatomy 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 239000012528 membrane Substances 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
- 238000013421 nuclear magnetic resonance imaging Methods 0.000 description 1
- 230000001575 pathological effect Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30016—Brain
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30096—Tumor; Lesion
Abstract
The invention relates to the technical field of image processing, in particular to a threshold segmentation method of brain tumor CT images, which comprises the following steps: obtaining a segmented image of the current CT image after fixed threshold segmentation processing and marking each existing connected region; calculating and obtaining a target communication area by combining with the morphological characteristics of the tumor; marking and storing the target communication area; reasonably and accurately selecting the position of seed point throwing according to the target communication area; and growing to obtain a segmentation effect image. The invention combines morphological characteristics of tumors and other intracranial tissues; the method has the advantages that the accuracy of the segmented image obtained by the region growth segmentation method is higher and the effect is better due to the self-adaptive evaluation and the selection of accurate regions for seed point throwing.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a threshold segmentation method of brain tumor CT images.
Background
Brain tumors refer to tumors formed in brain tissue or under the brain membrane, and one of the main methods for early screening of brain tumors is nuclear magnetic resonance imaging detection; the CT image of brain tumor needs to be subjected to threshold segmentation; because the traditional segmentation method is a fixed threshold method, namely, a fixed threshold is applied to the whole target image, and the tumor and the normal area are distinguished by determining the threshold, the method is simple and easy, but the obtained segmented image is often not up to the expected segmentation effect due to the influence of noise and ambient brightness.
Therefore, the current segmentation mode of tumor CT images selects an area growth segmentation algorithm with weak sensitivity to noise and ambient illumination brightness. Although the segmentation method of the region growing can overcome the sensitivity of the fixed threshold method to noise, the segmentation method is sensitive to the selection of initial seed points, and because some tissue regions such as blood vessels, gland tissues and the like in the brain are displayed with similar gray values in the image in the tumor region, the segmentation method can cause interference to the throwing of the seed points, and the phenomena of overgrowth and misclassification of the regions are easy to occur.
Disclosure of Invention
The invention provides a threshold segmentation method of brain tumor CT images, which aims to solve the existing problems.
The threshold segmentation method of the brain tumor CT image adopts the following technical scheme:
an embodiment of the present invention provides a method for threshold segmentation of brain tumor CT images, the method comprising the steps of:
acquiring all connected areas of the CT image in the segmented image after threshold segmentation;
obtaining a suspected tumor degree value of the communication region according to the difference value of the maximum value and the minimum value of the distances from the edge pixel points of the communication region to the mass center; a suspected tumor communicating region is obtained according to the suspected tumor degree value of the communicating region; obtaining a first credibility according to the duty ratio of the high gray pixel points in the suspected tumor connected region and the gray value average value; obtaining a first reliable high-communication region according to the first reliability degree of all the suspected tumor communication regions; obtaining a second credibility according to the gradation degree of the gray value in the first credibility high-communication area; obtaining a target communication area according to the second credibility of all the first credible high communication areas;
marking and storing the target communication area;
selecting a seed point throwing position according to the target communication area; and growing to obtain a segmentation result image.
Preferably, the step of acquiring all connected regions existing in the segmented image after the threshold segmentation process includes the following specific steps:
processing the CT image by using a threshold segmentation method to obtain a segmented effect image, and performing morphological open operation on the segmented image; and marking all connected areas in the image after the open operation processing, and further obtaining all connected areas in the segmented image after the threshold segmentation processing of the CT image.
Preferably, the obtaining the suspected tumor degree value of the connected region according to the difference value between the maximum value and the minimum value of the distance between the edge pixel point of the connected region and the centroid comprises the following specific steps:
obtaining the edges of all the communication areas by using a canny edge detection technology, and obtaining the mass centers of all the communication areas through images; analyzing any one of the communication areas, and marking the communication area as the current communication area; the expression for the suspected tumor extent value is:
；
in the method, in the process of the invention,a value representing a suspected tumor extent; />Centroid point representing current connected region, +.>Respectively representing the horizontal coordinate and the vertical coordinate values of the pixel points of the centroid of the current communication area in the image; />Representing edge pixel points; />On the edge representing the current connected region +.>A plurality of pixel points; />Indicating the current +.>An abscissa value and an ordinate value of each edge pixel point in the image; />Representing the maximum value of Euclidean distances from all edge pixel points of the current connected region to the centroid of the current connected region;a minimum value representing the Euclidean distance from all edge pixel points of the current connected region to the centroid thereof; />An exponential function based on a natural constant is represented.
Preferably, the obtaining the first confidence level according to the high gray pixel point duty ratio and the gray value mean value in the suspected tumor connected region includes the following specific steps:
analyzing any suspected tumor connected region and marking the analyzed suspected tumor connected region as a current suspected tumor connected region; the calculation expression of the first credibility of the current suspected tumor connected region is as follows:
；
in the method, in the process of the invention,a first confidence level indicating the current suspected tumor connected region,/for>Representing the number of pixel points contained in the current suspected tumor connected region, < >>Representing gray values +.>Represents any +.about.within the current suspected tumor connected region>Pixels>Then is whenArbitrary +.>Gray value of each pixel, +.>For the preset weight size, < ->Representing the maximum gray value of pixel points in the current suspected tumor connected region, < >>Then the minimum gray value +.>Indicating the +.f in the current suspected tumor connected region>Gray values of the individual pixels; />Indicating that the gray value in the current suspected tumor connected region is larger than the middle threshold value +.>The number of pixels of +.>Is->A function.
Preferably, the obtaining the second confidence level according to the gray value gradual change degree in the first confidence high-connectivity area includes the following specific steps:
analyzing any one of the first reliable high-communication areas, and recording the analyzed first reliable high-communication area as the current first reliable high-communication area, wherein a calculation expression of the second reliable degree of the current first reliable high-communication area is as follows:
；
in the method, in the process of the invention,representing a second confidence level value; />Representing the number of pixel points; />A centroid coordinate point of the communication area with high current first credibility is represented; />A third value representing the downward rounding of the average distance from the centroid to the edge point of the current first highly-trusted connected region; />Representing a currently first highly trusted connected regionIs used as the center of a circle and is>All pixel points contained in a circular range with radius; />Representing a gray value; />Representing the>Gray values of the individual pixels; />Representation->The average value of gray values of all pixel points in the corresponding circular range; />An exponential function that is based on a natural constant;representing the currently first communicating area with high confidence level to +.>Is used as the center of a circle and is>All pixel points contained in a circular range with radius; />Representing a currently first highly trusted connected regionIs used as the center of a circle and is>All pixel points contained in a circular range with radius;representation->The average value of gray values of all pixel points in the corresponding circular range; />Representation->The average value of gray values of all pixel points in the corresponding circular range.
Preferably, the selecting the position of the seed point according to the target communication area includes the following specific steps:
for the obtained target communication area; selecting the seed point as an optimal seed point throwing area, thereby realizing a throwing area of the seed point; a segmented image obtained using a region-growing segmentation method.
The technical scheme of the invention has the beneficial effects that: the traditional region growing algorithm is sensitive to the selection of initial seed points, and because partial tissue regions in the brain in a scene show similar gray values in the tumor regions in the images, the method can cause interference to the throwing of the seed points, and the phenomena of overgrowth and misclassification of the regions are easy to occur; the scheme combines morphological characteristics of tumors and other intracranial tissues; the method has the advantages that the accuracy of the segmented image obtained by the region growth segmentation method is higher and the effect is better due to the self-adaptive evaluation and the selection of accurate regions for seed point throwing.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart illustrating steps of a method for thresholding a brain tumor CT image according to the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the present invention to achieve the preset purposes, the following detailed description refers to specific embodiments, structures, features and effects of a method for segmenting brain tumor CT images 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 following specifically describes a specific scheme of the threshold segmentation method for brain tumor CT images provided by the present invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for thresholding brain tumor CT images according to an embodiment of the present invention is shown, the method includes the following steps:
step S001: and obtaining a segmented image of the CT image after threshold segmentation processing and marking each existing connected region.
In CT images, since a tissue region such as a blood vessel or a gland contained in the brain is similar in gray value to a tumor region in the image, a fixed threshold method is first used to preprocess the image to obtain a roughly divided image; the method is easy to divide into the same class, namely, the same class is divided into effect images (including the influence of the environment such as noise generated in the current process); therefore, preprocessing operation is needed to be carried out on the images obtained by rough segmentation, and all connected domains existing in the images are obtained, so that subsequent judgment and recognition can be conveniently carried out.
Specifically, firstly, a gray histogram of a current CT image is analyzed, a peak value is selected as a threshold value of a threshold segmentation method, the current CT image is processed by the threshold segmentation method to obtain a segmented effect image, and then the segmented image is subjected to preprocessing operation by morphological opening operation, so that the shape boundary of each connected domain can be smoothed by firstly corroding and expanding the image, isolated tiny pixel points and burrs existing in segmentation in the image are removed, and the overall position morphology of the connected domain in the image is not changed; on the basis of not changing the characteristic form, the interference caused by redundant pixel points and other factors is reduced; marking all connected regions in the image after the open operation processing, and sequentially storing the connected regions in a connected region setMiddle: -at->。
Thus, all connected areas in the segmented image of the CT image after the fixed threshold segmentation processing are obtained.
Step S002: and calculating and obtaining a target communication area by combining the morphological characteristics of the tumor.
It should be noted that the morphological characteristics of brain tumors have a certain diversity, and the specific manifestations are different according to factors such as tumor types, differentiation degrees, distribution positions and the like; thus it cannot be distinguished from other glandular tissues of the brain by only one feature; therefore, it is necessary to find out the multiple commonalities of the tumor under the influence of the diversified characteristics of the tumor, and the common characteristics of the current tumor can generate dissimilarity with other gland tissues, namely dissimilarity; first, many brain tumors are generally spherical or nearly spherical; the appearance characteristics can be used as the better distinguishing basis of tumors and other tissues; although the benign and malignant tumors have different clear and smooth edges, the benign and malignant tumors cannot be accurately distinguished through edge characteristics; but because both have higher densities inside; and the density change rule is similar in the interior; thus this feature can be an internal common feature of the tumor region and is distinguished from normal glandular tissue; and the connected domain is analyzed and calculated by combining the common characteristics of the tumor under the diversity characteristics, so that the tumor target connected domain is finally obtained.
1. And obtaining a suspected tumor communication region according to the external appearance morphological characteristics of the tumor.
It should be noted that brain tumors usually have a spherical or oval shape, while other brain intracranial tissue morphological features have a larger difference from them, mostly a narrow and long shape or an irregular shape; therefore, the connected domains are firstly subjected to preliminary analysis and judgment according to the appearance characteristics of the connected domains, so that the connected domains similar to the brain tumor form are obtained, and the suspected tumor connected domains are obtained.
Specifically, a canny edge detection technology is used for obtaining the edges of all the communication areas, and the mass centers of all the communication areas are obtained through images; analyzing any one of the communication areas, and marking the communication area as the current communication area; the suspected tumor extent assessment model is:
；
in the middle of，A value representing a suspected tumor extent; />Centroid point representing current connected region, +.>Respectively representing the horizontal coordinate and the vertical coordinate values of the pixel points of the centroid of the current communication area in the image; />Representing edge pixel points; />On the edge representing the current connected region +.>Pixels>Indicating the current +.>The abscissa value and the ordinate value of each edge pixel point in the image, and the data in the root numbers of the two polynomials are the +.>Euclidean distances of the edge pixels to their centroid; the first term is the maximum value of the calculation result, the second term is the minimum value of the calculation result, and the smaller the difference value between the first term and the second term is, the more circular the form of the current communication area is, the more regular the form is;representing the maximum value of Euclidean distances from all edge pixel points of the current connected region to the centroid of the current connected region; />All edges representing the current connected regionA minimum value of Euclidean distance from a pixel point to its centroid;
an exponential function that is based on a natural constant; mapping the final calculation result to a value range (0-1)]Between them; and the smaller the absolute value of the difference value is, the closer the obtained suspected tumor degree value is to 1. In this embodiment, setting the suspected tumor degree value to 0.7 is described, and the suspected tumor degree value may be set in combination with a specific scene during implementation; when->When the current communication region is considered to be a suspected tumor region, the degree value is high; marking the current connected domain as a suspected tumor connected domain, deleting the connected domain which does not meet the threshold value requirement, and not carrying out subsequent related treatment.
And calculating all the connected areas according to the suspected tumor degree evaluation model to obtain the suspected tumor connected areas.
2. And obtaining a first credibility evaluation model according to the internal characteristics of the analysis suspected tumor connected region.
It should be noted that, in the above steps, the morphological characteristics of each connected domain are analyzed to obtain a connected domain in which a suspected tumor region is obtained, and in the current step, the suspected tumor region is further analyzed and judged according to the internal characteristics, and as most of tumor cell kernels of each type are larger and more closely arranged than normal gland tissues in brain cranium, the tumor cell kernels generally have higher density, i.e. are represented in CT images to have higher gray values; because of the pathological characteristics of the tumor and the random diffusivity of the cancer cell distribution, the density change is not uniform and similar, and generally, the gray value of the region in the center of the tumor region is higher, and the density of the region which diverges outwards is relatively lower, namely the gray value is lower; and deeply analyzing the interior of the tumor by utilizing the characteristics to obtain a credibility evaluation model of the suspected tumor area.
Specifically, any one suspected tumor communication region is analyzed and marked as the current suspected tumor communication region; the first confidence level assessment model is:
；
in the method, in the process of the invention,a first confidence level value representing the current suspected tumor connected region,>representing the number of pixel points contained in the current suspected tumor connected region, < >>Representing gray values +.>Represents the +.sup.th in the current suspected tumor connected region>Pixels>Then is the +.f in the current suspected tumor connected region>Gray value of each pixel, +.>Is->Function, playing a normalization role, and mapping the gray value mean value calculated in the current bracket to a value range [ 0-1 ]]Between them; and the larger the satisfied value is, the closer the normalized result is to 1; />Representing the maximum gray value of pixel points in the current suspected tumor connected region, < >>Then the minimum gray value +.>Indicating the +.f in the current suspected tumor connected region>Gray values of the individual pixels;indicating that the gray value in the current suspected tumor connected region is larger than the middle threshold value +.>The number of pixels;
for the preset weight, the first term is used for calculating the gray value of the connected region, the size of the gray value can intuitively reflect the high-low state of density, and the second term is used for reflecting the gray value of more pixels in the current connected region, the larger the value is, the larger the gray value of the pixels in the current connected region is, and the higher the similarity of morphological characteristics in tumors is; thus give weight +.>The method comprises the steps of carrying out a first treatment on the surface of the The larger the final first confidence level value is, the larger the probability that the corresponding communication region is a suspected tumor region is; in this embodiment, the setting of the first confidence level value to 0.7 is described, and the first confidence level value may be set in association with a specific scene during implementation; when meeting->The first confidence level of the suspected tumor connected region is considered to be high when this inequality is given.
And calculating all the suspected tumor connected regions according to the first credibility assessment model to obtain a connected region with high credibility.
3. And obtaining a second credibility evaluation model according to the internal gray value change rule of the communication area with high credibility.
Specifically, any one of the connected regions with high first credibility is analyzed and recorded as the connected region with high first credibility at present, and then the acquired second credibility evaluation model of the connected region with high first credibility is:
；
in the method, in the process of the invention,representing a second confidence level value; />Representing the number of pixel points; />A centroid coordinate point of the communication area with high current first credibility is represented; />A third value representing the downward rounding of the average distance from the centroid to the edge point of the current first highly-trusted connected region; />Representing a currently first highly trusted connected regionIs the center of a circle>All pixel points contained in a circular range with radius; />Representing a gray value; />Representing the>Gray values of the individual pixels; />Representation->The average value of gray values of all pixel points in the corresponding circular range; according to the change rule of the gray value in the communication area with high current first credibility, when the ratio is closer to 1, the second credibility of the communication area with high current first credibility is higher, and then the result value in the absolute value is closer to 0, and the second credibility is higher; />An exponential function that is based on a natural constant; />Representing the currently first communicating area with high confidence level to +.>Is used as the center of a circle and is>All pixel points contained in a circular range with radius; />Representing the currently first communicating area with high confidence level to +.>Is used as the center of a circle and is>All pixel points contained in a circular range with radius;representation->The average value of gray values of all pixel points in the corresponding circular range; />Representation->The average value of gray values of all pixel points in the corresponding circular range. Mapping the final calculation result to a value range [ 0-1 ]]Between them; and the smaller the absolute value of the difference value is, the closer to 1 the obtained second confidence level value is. In this embodiment, setting the second confidence level value to 0.8 is described, and the second confidence level value may be set in association with a specific scene during implementation; when->When in use; the second credibility of the communication area with the current first credibility is considered to be high; the communication areas with high current first reliability are divided into a target communication area and a communication area with low reliability according to a threshold value.
And calculating all the communication areas with high first credibility according to the second credibility evaluation model to obtain the target communication area.
Step S003: and marking and storing the target communication area.
Specifically, according to the three evaluation models, each connected domain is calculated to obtain a connected domain with high second reliability, and the connected domain with high second reliability is finally marked and stored as a target connected domain.
Step S004: selecting a seed point throwing position according to the target communication area; and obtaining a segmentation result image by using a region growing segmentation method.
Specifically, for the obtained target communication region; selecting the seed point as an optimal seed point throwing area, wherein other communication areas are not thrown, so that a throwing area of the seed point is realized; and dividing the CT image by using a region growing dividing method to obtain a dividing result image.
The segmented image obtained through the steps has higher accuracy and better effect.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (4)
1. A method for threshold segmentation of brain tumor CT images, comprising the steps of:
acquiring all connected areas of the CT image in the segmented image after threshold segmentation;
obtaining a suspected tumor degree value of the communication region according to the difference value of the maximum value and the minimum value of the distances from the edge pixel points of the communication region to the mass center; a suspected tumor communicating region is obtained according to the suspected tumor degree value of the communicating region; obtaining a first credibility according to the duty ratio of the high gray pixel points in the suspected tumor connected region and the gray value average value; obtaining a first reliable high-communication region according to the first reliability degree of all the suspected tumor communication regions; obtaining a second credibility according to the gradation degree of the gray value in the first credibility high-communication area; obtaining a target communication area according to the second credibility of all the first credible high communication areas;
marking and storing the target communication area;
selecting a seed point throwing position according to the target communication area; and growing to obtain a segmentation result image;
the method for obtaining the first credibility according to the high gray pixel point duty ratio and the gray value mean value in the suspected tumor connected region comprises the following specific steps:
analyzing any suspected tumor connected region and marking the analyzed suspected tumor connected region as a current suspected tumor connected region; the calculation expression of the first credibility of the current suspected tumor connected region is as follows:
in the method, in the process of the invention,a first confidence level indicating the current suspected tumor connected region,/for>Representing the number of pixel points contained in the current suspected tumor connected region, < >>Representing gray values +.>Represents any +.about.within the current suspected tumor connected region>Pixels>Then be any +.>Gray value of each pixel, +.>For the preset weight size, < ->Representing the maximum gray value of pixel points in the current suspected tumor connected region, < >>Then the minimum gray value +.>Indicating the +.f in the current suspected tumor connected region>Gray values of the individual pixels; />Indicating that the gray value in the current suspected tumor connected region is larger than the middle threshold value +.>Is used for displaying the number of the pixel points,a function;
the second credibility level is obtained according to the gradation level of the gray value in the first credibility high-communication area, and the method comprises the following specific steps:
analyzing any one of the first reliable high-communication areas, and recording the analyzed first reliable high-communication area as the current first reliable high-communication area, wherein a calculation expression of the second reliable degree of the current first reliable high-communication area is as follows:
in the method, in the process of the invention,representing a second confidence level value; />Representing the number of pixel points; />A centroid coordinate point of the communication area with high current first credibility is represented; />A third value representing the downward rounding of the average distance from the centroid to the edge point of the current first highly-trusted connected region; />Representing a current first high level of confidenceThrough areaIs used as the center of a circle and is>All pixel points contained in a circular range with radius; />Representing a gray value; />Representing the>Gray values of the individual pixels; />Representation->The average value of gray values of all pixel points in the corresponding circular range; />An exponential function that is based on a natural constant;representing the currently first communicating area with high confidence level to +.>Is used as the center of a circle and is>All pixel points contained in a circular range with radius; />Representing a currently first highly trusted connected regionTo->Is used as the center of a circle and is>All pixel points contained in a circular range with radius;representation->The average value of gray values of all pixel points in the corresponding circular range; />Representation->The average value of gray values of all pixel points in the corresponding circular range.
2. The method for threshold segmentation of brain tumor CT images according to claim 1, wherein the step of acquiring all connected regions of the CT images in the segmented image after the threshold segmentation process comprises the following specific steps:
processing the CT image by using a threshold segmentation method to obtain a segmented effect image, and performing morphological open operation on the segmented image; and marking all connected areas in the image after the open operation processing, and further obtaining all connected areas in the segmented image after the threshold segmentation processing of the CT image.
3. The method for threshold segmentation of brain tumor CT images according to claim 1, wherein the obtaining the suspected tumor degree value of the connected region according to the difference between the maximum value and the minimum value of the distances from the edge pixel points of the connected region to the centroid comprises the following specific steps:
obtaining the edges of all the communication areas by using a canny edge detection technology, and obtaining the mass centers of all the communication areas through images; analyzing any one of the communication areas, and marking the communication area as the current communication area; the expression for the suspected tumor extent value is:
in the method, in the process of the invention,a value representing a suspected tumor extent; />Centroid point representing current connected region, +.>Respectively representing the horizontal coordinate and the vertical coordinate values of the pixel points of the centroid of the current communication area in the image; />Representing edge pixel points; />On the edge representing the current connected region +.>A plurality of pixel points; />Indicating the current +.>An abscissa value and an ordinate value of each edge pixel point in the image; />Representing all edge pixel points of the current connected region to the centroid thereofA maximum value of euclidean distance;a minimum value representing the Euclidean distance from all edge pixel points of the current connected region to the centroid thereof; />An exponential function based on a natural constant is represented.
4. The method for threshold segmentation of brain tumor CT images according to claim 1, wherein the selecting the position of the seed point according to the target communication region comprises the following specific steps:
obtaining a target communication area; selecting the seed point as an optimal seed point throwing area, thereby realizing a throwing area of the seed point; segmentation result images obtained using the region growing segmentation method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310912887.0A CN116630358B (en) | 2023-07-25 | 2023-07-25 | Threshold segmentation method for brain tumor CT image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310912887.0A CN116630358B (en) | 2023-07-25 | 2023-07-25 | Threshold segmentation method for brain tumor CT image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116630358A CN116630358A (en) | 2023-08-22 |
| CN116630358B true CN116630358B (en) | 2023-09-26 |
Family
ID=87590673
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310912887.0A Active CN116630358B (en) | 2023-07-25 | 2023-07-25 | Threshold segmentation method for brain tumor CT image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116630358B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116912246B (en) * | 2023-09-13 | 2023-12-29 | 潍坊医学院 | Tumor CT data processing method based on big data |
| CN117237389B (en) * | 2023-11-14 | 2024-01-19 | 深圳市亿康医疗技术有限公司 | CT image segmentation method for middle ear cholesteatoma |
Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2339797A1 (en) * | 1998-08-07 | 2000-03-02 | Arch Development Corporation | Method and system for the segmentation and classification of lesions |
| US6246784B1 (en) * | 1997-08-19 | 2001-06-12 | The United States Of America As Represented By The Department Of Health And Human Services | Method for segmenting medical images and detecting surface anomalies in anatomical structures |
| CN1395713A (en) * | 2000-01-18 | 2003-02-05 | 芝加哥大学 | Method, system and computer readable medium for two-dimensional and three-dimensional detection of lungs nodules in computed tomography image scans |
| CN102360495A (en) * | 2011-10-19 | 2012-02-22 | 西安电子科技大学 | Pulmonary nodule segmentation method based on average intensity projection and translation gaussian model |
| CN103236062A (en) * | 2013-05-03 | 2013-08-07 | 北京国铁华晨通信信息技术有限公司 | Magnetic resonance imaging blood vessel segmentation method and system based on human brain tumor nuclear magnetic library |
| CN104091347A (en) * | 2014-07-26 | 2014-10-08 | 刘宇清 | Intracranial tumor operation planning and simulating method based on 3D print technology |
| WO2015067300A1 (en) * | 2013-11-05 | 2015-05-14 | Brainlab Ag | Determination of enhancing structures in an anatomical body part |
| CN105488781A (en) * | 2015-06-01 | 2016-04-13 | 深圳市第二人民医院 | Dividing method based on CT image liver tumor focus |
| WO2017019059A1 (en) * | 2015-07-29 | 2017-02-02 | Perkinelmer Health Sciences, Inc. | Systems and methods for automated segmentation of individual skeletal bones in 3d anatomical images |
| CN106780518A (en) * | 2017-02-10 | 2017-05-31 | 苏州大学 | A kind of MR image three-dimensional interactive segmentation methods of the movable contour model cut based on random walk and figure |
| WO2021146705A1 (en) * | 2020-01-19 | 2021-07-22 | Ventana Medical Systems, Inc. | Non-tumor segmentation to support tumor detection and analysis |
| CN113711271A (en) * | 2019-03-15 | 2021-11-26 | 豪夫迈·罗氏有限公司 | Deep convolutional neural network for tumor segmentation by positron emission tomography |
| CN114764809A (en) * | 2021-01-12 | 2022-07-19 | 电科云(北京)科技有限公司 | Self-adaptive threshold segmentation method and device for lung CT (computed tomography) density increase shadow |
| CN115311294A (en) * | 2022-10-12 | 2022-11-08 | 启东金耀億华玻纤材料有限公司 | Glass bottle body flaw identification and detection method based on image processing |
| CN115330800A (en) * | 2022-10-14 | 2022-11-11 | 深圳市亿康医疗技术有限公司 | Automatic segmentation method for radiotherapy target area based on image processing |
| CN115797356A (en) * | 2023-02-09 | 2023-03-14 | 山东第一医科大学附属省立医院(山东省立医院) | Nuclear magnetic resonance tumor region extraction method |
| CN115841472A (en) * | 2022-12-07 | 2023-03-24 | 沈阳东软智能医疗科技研究院有限公司 | Method, device, equipment and storage medium for identifying high-density characteristics of middle cerebral artery |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11350901B2 (en) * | 2017-11-27 | 2022-06-07 | Case Western Reserve University | Recurrence prognosis and prediction of added benefit of adjuvant chemotherapy in early stage non-small cell lung cancer with radiomic features on baseline computed tomography |
| US11227387B2 (en) * | 2020-05-18 | 2022-01-18 | Prince Mohammad Bin Fahd University | Multi-stage brain tumor image processing method and system |
-
2023
- 2023-07-25 CN CN202310912887.0A patent/CN116630358B/en active Active
Patent Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6246784B1 (en) * | 1997-08-19 | 2001-06-12 | The United States Of America As Represented By The Department Of Health And Human Services | Method for segmenting medical images and detecting surface anomalies in anatomical structures |
| CA2339797A1 (en) * | 1998-08-07 | 2000-03-02 | Arch Development Corporation | Method and system for the segmentation and classification of lesions |
| CN1395713A (en) * | 2000-01-18 | 2003-02-05 | 芝加哥大学 | Method, system and computer readable medium for two-dimensional and three-dimensional detection of lungs nodules in computed tomography image scans |
| CN102360495A (en) * | 2011-10-19 | 2012-02-22 | 西安电子科技大学 | Pulmonary nodule segmentation method based on average intensity projection and translation gaussian model |
| CN103236062A (en) * | 2013-05-03 | 2013-08-07 | 北京国铁华晨通信信息技术有限公司 | Magnetic resonance imaging blood vessel segmentation method and system based on human brain tumor nuclear magnetic library |
| WO2015067300A1 (en) * | 2013-11-05 | 2015-05-14 | Brainlab Ag | Determination of enhancing structures in an anatomical body part |
| CN104091347A (en) * | 2014-07-26 | 2014-10-08 | 刘宇清 | Intracranial tumor operation planning and simulating method based on 3D print technology |
| CN105488781A (en) * | 2015-06-01 | 2016-04-13 | 深圳市第二人民医院 | Dividing method based on CT image liver tumor focus |
| WO2017019059A1 (en) * | 2015-07-29 | 2017-02-02 | Perkinelmer Health Sciences, Inc. | Systems and methods for automated segmentation of individual skeletal bones in 3d anatomical images |
| CN106780518A (en) * | 2017-02-10 | 2017-05-31 | 苏州大学 | A kind of MR image three-dimensional interactive segmentation methods of the movable contour model cut based on random walk and figure |
| CN113711271A (en) * | 2019-03-15 | 2021-11-26 | 豪夫迈·罗氏有限公司 | Deep convolutional neural network for tumor segmentation by positron emission tomography |
| WO2021146705A1 (en) * | 2020-01-19 | 2021-07-22 | Ventana Medical Systems, Inc. | Non-tumor segmentation to support tumor detection and analysis |
| CN114764809A (en) * | 2021-01-12 | 2022-07-19 | 电科云(北京)科技有限公司 | Self-adaptive threshold segmentation method and device for lung CT (computed tomography) density increase shadow |
| CN115311294A (en) * | 2022-10-12 | 2022-11-08 | 启东金耀億华玻纤材料有限公司 | Glass bottle body flaw identification and detection method based on image processing |
| CN115330800A (en) * | 2022-10-14 | 2022-11-11 | 深圳市亿康医疗技术有限公司 | Automatic segmentation method for radiotherapy target area based on image processing |
| CN115841472A (en) * | 2022-12-07 | 2023-03-24 | 沈阳东软智能医疗科技研究院有限公司 | Method, device, equipment and storage medium for identifying high-density characteristics of middle cerebral artery |
| CN115797356A (en) * | 2023-02-09 | 2023-03-14 | 山东第一医科大学附属省立医院(山东省立医院) | Nuclear magnetic resonance tumor region extraction method |
Non-Patent Citations (1)
| Title |
|---|
| A smoke segmentation algorithm based on improved intelligent seeded region growing;Zhao, WD等;《FIRE AND MATERIALS》;第43卷(第6期);725-733 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116630358A (en) | 2023-08-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116630358B (en) | Threshold segmentation method for brain tumor CT image | |
| US7736313B2 (en) | Detecting and classifying lesions in ultrasound images | |
| WO2021129323A1 (en) | Ultrasound image lesion describing method and apparatus, computer device, and storage medium | |
| CN115330800B (en) | Automatic segmentation method of radiotherapy target area based on image processing | |
| CN108830853A (en) | A kind of melanoma aided diagnosis method based on artificial intelligence | |
| CN115359052B (en) | Medical image enhancement method based on clustering algorithm | |
| WO2008069762A1 (en) | Method for identifying a pathological region of a scan, such as an ischemic stroke region of an mri scan | |
| CN107292312A (en) | Tumour recognition methods | |
| CN112241948A (en) | Lung nodule detection and analysis method and system capable of adapting to thickness of layer | |
| CN116912255B (en) | Follicular region segmentation method for ovarian tissue analysis | |
| CN106529188A (en) | Image processing method applied to surgical navigation | |
| CN117237591A (en) | Intelligent removal method for heart ultrasonic image artifacts | |
| CN116485819B (en) | Ear-nose-throat examination image segmentation method and system | |
| CN112419282A (en) | Automatic detection method and system for aneurysm in brain medical image | |
| CN113924046A (en) | Computer-based method for classifying organ masses as cysts | |
| CN113409275B (en) | Method for determining thickness of transparent layer behind fetal neck based on ultrasonic image and related device | |
| CN113191413B (en) | Prostate multimode MR image classification method and system based on foveal residual error network | |
| CN112950734A (en) | Coronary artery reconstruction method, device, electronic equipment and storage medium | |
| Tivatansakul et al. | Breast mass detection from mammography using iteration of gray-level co-occurrence matrix | |
| CN117237389B (en) | CT image segmentation method for middle ear cholesteatoma | |
| Song et al. | A robust fuzzy c-means clustering model with spatial constraint for brain magnetic resonance image segmentation | |
| CN117593192B (en) | Gynecological cervical image enhancement analysis method | |
| CN117575977B (en) | Follicular region enhancement method for ovarian tissue analysis | |
| CN117252893B (en) | Segmentation processing method for breast cancer pathological image | |
| CN112258535B (en) | Integrated positioning and segmentation method for corpus callosum and lumbricus in ultrasonic image |
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 |