CN116485814A - Intracranial hematoma region segmentation method based on CT image - Google Patents
Intracranial hematoma region segmentation method based on CT image Download PDFInfo
- Publication number
- CN116485814A CN116485814A CN202310460604.3A CN202310460604A CN116485814A CN 116485814 A CN116485814 A CN 116485814A CN 202310460604 A CN202310460604 A CN 202310460604A CN 116485814 A CN116485814 A CN 116485814A
- Authority
- CN
- China
- Prior art keywords
- area
- irregular
- hematoma
- pixel points
- pixel point
- 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.)
- Pending
Links
- 206010059491 Intracranial haematoma Diseases 0.000 title claims abstract description 39
- 238000000034 method Methods 0.000 title claims abstract description 25
- 230000011218 segmentation Effects 0.000 title claims abstract description 24
- 230000001788 irregular Effects 0.000 claims abstract description 127
- 206010018852 Haematoma Diseases 0.000 claims abstract description 61
- 238000007917 intracranial administration Methods 0.000 claims abstract description 22
- 238000003708 edge detection Methods 0.000 claims description 3
- 238000003709 image segmentation Methods 0.000 abstract description 7
- 230000000694 effects Effects 0.000 description 8
- 210000004556 brain Anatomy 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 210000005013 brain tissue Anatomy 0.000 description 2
- 210000003625 skull Anatomy 0.000 description 2
- 201000002345 Brain compression Diseases 0.000 description 1
- 206010018985 Haemorrhage intracranial Diseases 0.000 description 1
- 208000008574 Intracranial Hemorrhages Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 210000004204 blood vessel Anatomy 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008733 trauma Effects 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
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/181—Segmentation; Edge detection involving edge growing; involving edge linking
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20112—Image segmentation details
- G06T2207/20152—Watershed segmentation
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The invention relates to the technical field of image processing, in particular to a CT image-based intracranial hematoma region segmentation method, which comprises the following steps: acquiring intracranial CT images, further obtaining irregular areas in the segmented images, obtaining edge pixel points in each irregular area, obtaining a target degree value of each irregular area according to a gray average value and the edge pixel points of each irregular area, taking the irregular area with the maximum target degree value as a starting area, calculating a combined probability value between the starting area and the adjacent irregular area from the starting area, traversing and combining the irregular areas to obtain a suspected hematoma area, and correcting the suspected hematoma area according to contour pixel points in the suspected hematoma area and common pixel points of the contour pixel points to obtain the intracranial hematoma area. The intracranial hematoma area obtained by the method is accurate and complete, and compared with the existing image segmentation algorithm, the accuracy is further improved.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a CT image-based intracranial hematoma region segmentation method.
Background
Intracranial hematoma is a condition of intracranial hemorrhage caused by compression of brain tissue due to rupture of blood vessels in the brain or between brain tissue and skull caused by trauma and the like, and blood is accumulated in the brain or between the brain and skull. The hematoma area needs to be segmented from the intracranial CT image firstly for research, and the medical judgment is affected by the segmentation effect, so that a method for precisely segmenting the hematoma area in the intracranial CT image is needed.
Because of complex gray level distribution of intracranial CT images, the existing image segmentation algorithm such as threshold segmentation can be interfered by gray levels of other areas, and hematoma areas can not be completely segmented. Because of noise and texture effects in the intracranial CT image, the existing image segmentation algorithm such as watershed segmentation algorithm can generate an excessive segmentation effect, so that the intracranial hematoma area is excessively segmented, and an accurate intracranial hematoma area cannot be acquired for focus analysis.
Disclosure of Invention
The invention provides a CT image-based intracranial hematoma region segmentation method, which aims to solve the existing problems.
The invention discloses a CT image-based intracranial hematoma region segmentation method, which adopts the following technical scheme:
one embodiment of the invention provides a method for dividing intracranial hematoma areas based on CT images, which comprises the following steps:
acquiring intracranial CT images; dividing the intracranial CT image by using a watershed segmentation algorithm to obtain a segmented image, and taking each region in the segmented image as an irregular region; edge detection is carried out on each irregular area, and edge pixel points in each irregular area are obtained; acquiring a target degree value of each irregular area according to the gray average value of each irregular area and the edge pixel points;
performing traversal merging on the irregular area, including:
s1: taking an irregular area with the maximum target degree value as a starting area; taking the initial region as the initial merging region;
s2: respectively taking all irregular areas adjacent to the initial area as areas to be selected; calculating a combined probability value between each region to be selected and the initial region; combining the to-be-selected region with the combination probability value larger than the preset combination threshold value with the combination region to which the initial region belongs; taking the region to be selected with the combination probability value larger than the preset combination threshold value as a new initial region;
s3: repeating S2 according to the new initial region, and stopping iteration until all irregular regions are not merged any more; taking the latest obtained combined area as a suspected hematoma area;
taking any outline pixel point of the suspected hematoma area as a pixel point to be measured, and carrying out outline adjustment on the pixel point to be measured, wherein the method comprises the following steps: acquiring a reference pixel point of a pixel point to be detected; acquiring all public pixel points according to the reference pixel points and the pixel points to be detected, and updating the suspected hematoma area according to all public pixel points and the pixel points to be detected;
and respectively taking each contour pixel point on the contour of the suspected hematoma area as a pixel point to be measured according to the clockwise direction, carrying out contour adjustment on each pixel point to be measured, and taking the finally obtained suspected hematoma area as an intracranial hematoma area to realize the segmentation of the intracranial hematoma area.
Preferably, the obtaining the target degree value of each irregular area according to the gray average value of each irregular area and the edge pixel point includes the following specific steps:
wherein a is k A target degree value for the kth irregular area; p is p k The gray average value of the kth irregular area; p is p max The maximum value of the gray average value of all the irregular areas; p's' k,i A gray average value of an ith irregular area adjacent to the kth irregular area; n is n k The number of all irregular areas adjacent to the kth irregular area; c k The number of edge pixel points in the kth irregular area; d, d k The average value of Euclidean distances between every two edge pixel points in the kth irregular area; exp () is an exponential function based on a natural constant.
Preferably, the calculating the combined probability value between each candidate region and the initial region includes the following specific steps:
wherein sigma j A combined probability value between the j-th candidate region and the initial region; a' j The target degree value of the j-th candidate area is the target degree value; a is a target degree value of an initial region; p is p max The maximum value of the gray average value of all the irregular areas;the average value of the gray average value of the jth area to be selected and the gray average value of the initial area; the absolute value symbol; exp () is an exponential function based on a natural constant.
Preferably, the step of obtaining the reference pixel point of the pixel point to be measured includes the following specific steps:
and acquiring two contour pixel points which are adjacent to the pixel point to be detected and are closest to the pixel point to be detected, and respectively taking the two contour pixel points as reference pixel points of the pixel point to be detected.
Preferably, the step of obtaining all the common pixels according to the reference pixel and the pixel to be detected includes the following specific steps:
and taking the pixel points which are simultaneously positioned in eight adjacent areas of the pixel points to be detected and the two reference pixel points as the common pixel point of the pixel points to be detected.
Preferably, the updating the suspected hematoma area according to all the public pixel points and the pixel points to be detected includes the following specific steps:
acquiring the pixel points to be detected and the gradient amplitude of each public pixel point, and if the gradient amplitude of the pixel points to be detected is greater than or equal to the gradient amplitude of each public pixel point, not updating the suspected hematoma area;
if the gradient amplitude of the pixel to be detected is smaller than the gradient amplitude of one or more public pixel points, the public pixel point with the largest gradient amplitude is obtained, when the public pixel point with the largest gradient amplitude does not belong to the suspected hematoma area, the public pixel point with the largest gradient amplitude is merged into the suspected hematoma area, and when the public pixel point with the largest gradient amplitude belongs to the suspected hematoma area, the pixel point to be detected is removed from the suspected hematoma area.
The technical scheme of the invention has the beneficial effects that: according to the method, watershed segmentation is carried out on an intracranial CT image to obtain irregular areas in a segmented image, edge pixel points in each irregular area are obtained, a target degree value of each irregular area is obtained according to a gray average value of each irregular area and the edge pixel points, the probability that the irregular area is a hematoma area is measured, the irregular area with the maximum target degree value is taken as an initial area, a merging probability value between the initial area and the adjacent irregular area is calculated from the initial area, the irregular areas are traversed and merged to obtain a suspected hematoma area, the suspected hematoma area is corrected according to outline pixel points in the suspected hematoma area and common pixel points of the outline pixel points, and the intracranial hematoma area is obtained, so that the outline of the intracranial hematoma area contains more detailed information. The intracranial hematoma area obtained by the method is accurate and complete, and compared with the existing image segmentation algorithm, the accuracy is further improved.
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 flow chart showing the steps of a CT image-based method for segmenting an intracranial hematoma region according to the present invention;
FIG. 2 is an intracranial CT image;
FIG. 3 is a segmented image;
FIG. 4 is a partial detail of a segmented image;
fig. 5 is a schematic diagram of a pixel to be tested, a reference pixel, and a common pixel.
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 refers to specific embodiments, structures, features and effects of an intracranial hematoma region segmentation method based on 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 method for dividing intracranial hematoma region based on CT image.
Referring to fig. 1, a flowchart of a method for segmenting an intracranial hematoma region based on CT images according to an embodiment of the invention is shown, the method includes the following steps:
s001, acquiring intracranial CT images.
And scanning the head of the patient by using a CT machine to acquire an intracranial CT image. An intracranial CT image in accordance with an embodiment of the present invention is shown in FIG. 2.
Thus, an intracranial CT image is acquired.
S002, obtaining a segmented image to obtain an irregular segmented region.
It should be noted that, because of the complex gray level distribution of the intracranial CT image, the existing image segmentation algorithm such as threshold segmentation is interfered by gray levels of other regions, and the hematoma region cannot be completely segmented. Because of noise and texture effects in the intracranial CT image, the existing image segmentation algorithm such as watershed segmentation algorithm can generate an excessive segmentation effect, so that the intracranial hematoma area is excessively segmented, and an accurate intracranial hematoma area cannot be acquired for focus analysis. Therefore, the embodiment of the invention analyzes and merges the excessive images obtained by dividing the watershed to obtain the accurate intracranial hematoma region.
In the embodiment of the invention, the intracranial CT image is segmented by using the traditional watershed segmentation algorithm to obtain a segmented image. See fig. 3 for segmented images.
It should be noted that, fig. 4 is an image obtained by amplifying local details of a segmented image, and it can be found that the conventional watershed segmentation algorithm segments the image into irregular areas that are too small, so that the foreground and the background of the image are excessively segmented, and the segmentation effect is poor. To solve the over-segmentation problem, it is necessary to mark irregular areas in the segmented image.
In the embodiment of the invention, a sliding block with the length and the width of 1 is arranged, the segmented image is traversed from left to right and from top to bottom, and each traversed irregular region is numbered according to the traversing sequence.
Thus far, the divided image and the irregular region in the divided image are acquired.
S003, obtaining a target degree value of each irregular block area.
It should be noted that, the hematoma area is shown to have a certain gray level gradual change phenomenon in the intracranial CT image, and the gray level characteristic of large gray level value, so the target degree value of the irregular area can be obtained according to the gray level characteristic of the irregular area in the segmented image, and the probability that the irregular area is the hematoma area can be measured, so that the irregular area corresponding to the maximum target degree value can be combined from the beginning to the periphery, and the accurate hematoma area can be obtained.
In the embodiment of the invention, edge detection is carried out on each irregular area, and edge pixel points in each irregular area are obtained. And taking the average value of the gray values of all the pixel points in each irregular area as the gray average value of each irregular area. In the embodiment of the invention, the pixel points on the outline of the irregular area are called outline pixel points, and if the outline pixel points of the other irregular area exist in the eight neighborhood of the partial outline pixel points of one irregular area, the two irregular areas are adjacent.
Obtaining a target degree value of each irregular area according to the gray average value of each irregular area and the edge pixel points:
wherein a is k A target degree value for the kth irregular area; p is p k The gray average value of the kth irregular area; p is p max For p, the maximum value of the gray average value of all the irregular areas, i.e. the maximum gray average value of all the irregular areas k Normalizing; p's' k,i A gray average value of an ith irregular area adjacent to the kth irregular area; n is n k The number of all irregular areas adjacent to the kth irregular area; c k The number of edge pixel points in the kth irregular area; d, d k The average value of Euclidean distances between every two edge pixel points in the kth irregular area; exp () is an exponential function based on a natural constant; when the gray average value p of the kth irregular region k The greater the target degree value a of the kth irregular area k The larger the conversely, the lower;for the average difference between the gray average value of the kth irregular area and the gray average value of all adjacent irregular areas, the average difference is used for measuring the effectiveness of the gray average value of the kth irregular area, the denominator is added with one to prevent the denominator from being 0, when the average difference is smaller, the probability that the kth irregular area is a non-discrete area is higher, the reliability of the gray average value of the kth irregular area is higher, and the target degree value a of the kth irregular area is higher k The larger the conversely, the lower; />Gradient factor representing the kth irregular area, when the number c of edge pixel points in the kth irregular area k The more pixels in the kth irregular area with larger gradients are indicated, the more uneven the gray distribution is,if the average value of Euclidean distances between every two edge pixel points in the kth irregular area is smaller, namely the average distance of the edge pixel points in the kth irregular area is smaller, the edge pixel points are denser, the gradient change in the kth irregular area is larger, the gray level in the kth irregular area is more uneven, and the gray level is obtained comprehensively>To prevent the calculation influence caused by the fact that the numerator and the denominator are 0, add one to the numerator and the denominator and add one to the denominator>Representing a gradient factor for reflecting the uniformity of the gray scale distribution in the kth irregular region, wherein the more the gradient factor is, the more uneven the gray scale distribution in the kth irregular region is, the more the gradient factor is in accordance with the gray scale gradient characteristic of the hematoma region, and the target degree value a of the kth irregular region is k The larger the conversely, the lower.
Thus, the target degree value of each irregular area is obtained.
S004, acquiring an initial region, and merging the irregular regions to obtain a suspected hematoma region.
The greater the target degree value of the irregular area, the more likely the irregular area is a hematoma area. In order to acquire an accurate hematoma area, it is necessary to acquire an irregular area which is most likely to be the hematoma area according to the target degree value, and traverse and merge from the irregular area to the surrounding irregular area is started.
In the embodiment of the invention, the method for traversing and merging the irregular areas comprises the following steps:
1. and acquiring an irregular area with the maximum target degree value as a starting area, and taking the starting area as the initial merging area.
2. And (3) marking the target degree value of the initial region as A, and acquiring all irregular regions adjacent to the initial region as candidate regions respectively. Calculating a combined probability value between each candidate region and the initial region:
wherein sigma j The combined probability value between the j-th candidate region and the initial region can reflect the combined trend degree of the two regions; a' j The target degree value of the j-th candidate area is the target degree value; a is a target degree value of an initial region; p is p max The maximum gray average value of all the irregular areas is the maximum gray average value of all the irregular areas;the average value of the gray average value of the jth area to be selected and the gray average value of the initial area; the absolute value symbol; exp () is an exponential function based on a natural constant; |a' j -a| is the absolute value of the difference between the target degree value of the jth candidate region and the target degree value of the starting region, when the absolute value of the difference is smaller, the jth candidate region is closer to the target degree value of the starting region, and at the moment, the greater the combined probability value between the jth candidate region and the starting region is, and conversely, the smaller is; />For the gray scale influence factors of the jth candidate region and the initial region, when the average value of the gray scale mean value of the jth candidate region and the gray scale mean value of the initial region is larger, the greater the possibility that the region after the jth candidate region is combined with the initial region is a hematoma region is, the jth candidate region and the initial region are more required to be combined, and at the moment, the greater the combining probability value between the jth candidate region and the initial region is, and conversely, the smaller the combining probability value is.
If the combination probability value between one to-be-selected area and the initial area is larger than the preset combination threshold value T, combining the to-be-selected area with the combination area to which the initial area belongs to obtain a new combination area, and taking the to-be-selected area as the new initial area. In the embodiment of the present invention, the preset merge threshold t=0.75, and in other embodiments, the practitioner may set the preset merge threshold according to the actual implementation situation.
And similarly, judging all the regions to be selected, and merging according to the merging probability value between each region to be selected and the initial region to obtain all the new initial regions.
3. And (3) repeating the step (2) according to the new initial region, and stopping iteration until all irregular regions are no longer merged. The latest combined region obtained is the possible hematoma region, and is called a suspected hematoma region.
Thus, the combination of the irregular areas is realized, and the suspected hematoma areas are obtained.
S005, correcting the suspected hematoma area to obtain the intracranial hematoma area.
In order to obtain more detailed contour information, an accurate and complete hematoma region is obtained, and the contour of the suspected hematoma region is corrected so that the contour is aligned to the gradient direction.
In the embodiment of the invention, the outline pixel points of the suspected hematoma area are obtained, the gradient amplitude of each pixel point in the intracranial CT image is obtained by utilizing a Sobel operator, and in other embodiments, the implementation personnel can select other image gradient algorithms.
Taking any outline pixel point of the suspected hematoma area as a pixel point to be measured, and carrying out outline adjustment on the pixel point to be measured, wherein the outline adjustment specifically comprises the following steps:
and acquiring two contour pixel points which are adjacent to the pixel point to be detected and are closest to the pixel point to be detected, and respectively taking the two contour pixel points as reference pixel points of the pixel point to be detected. If one pixel point in the intracranial CT image is simultaneously positioned in eight adjacent areas of the pixel point to be detected and the two reference pixel points, the pixel point is used as a common pixel point of the pixel point to be detected, and all the common pixel points of the pixel point to be detected are obtained. Fig. 5 shows four local positional relationships among a pixel to be measured, a reference pixel, and a common pixel, wherein a black circle in fig. 5 is the pixel to be measured, a gray circle is the reference pixel, and a hollow circle is the common pixel. Comparing the gradient amplitude of the pixel to be detected with that of each common pixel, and if the gradient amplitude of the pixel to be detected is greater than or equal to that of each common pixel, reserving the pixel to be detected. If the gradient amplitude of the pixel to be detected is smaller than that of one or more public pixel points, the public pixel point with the largest gradient amplitude is obtained, if the public pixel point with the largest gradient amplitude does not belong to the suspected hematoma area, the public pixel point with the largest gradient amplitude is merged into the suspected hematoma area, if the public pixel point with the largest gradient amplitude belongs to the suspected hematoma area, the pixel point to be detected is removed from the suspected hematoma area, and at the moment, the public pixel point with the largest gradient amplitude becomes the outline pixel point of the suspected hematoma area.
And respectively taking each contour pixel point on the contour of the suspected hematoma area as a pixel point to be measured according to the clockwise direction, and carrying out contour adjustment on each pixel point to be measured to obtain a final suspected hematoma area which is an accurate and complete intracranial hematoma area.
Thus, an intracranial hematoma area was obtained.
Through the above steps, the division of the intracranial hematoma region is completed.
According to the embodiment of the invention, the watershed segmentation is carried out on the intracranial CT image to obtain irregular areas in the segmented image, edge pixel points in each irregular area are obtained, a target degree value of each irregular area is obtained according to the gray average value of each irregular area and the edge pixel points, the target degree value is used for measuring the possibility that the irregular area is a hematoma area, the irregular area with the maximum target degree value is taken as an initial area, the combined probability value between the initial area and the adjacent irregular area is calculated from the initial area, the irregular areas are traversed and combined to obtain a suspected hematoma area, the suspected hematoma area is corrected according to the outline pixel points in the suspected hematoma area and the common pixel points of the outline pixel points, and the intracranial hematoma area is obtained, so that the outline of the intracranial hematoma area contains more detailed information. The intracranial hematoma area obtained by the method is accurate and complete, and compared with the existing image segmentation algorithm, the accuracy is further improved.
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 (6)
1. An intracranial hematoma region segmentation method based on CT images is characterized by comprising the following steps:
acquiring intracranial CT images; dividing the intracranial CT image by using a watershed segmentation algorithm to obtain a segmented image, and taking each region in the segmented image as an irregular region; edge detection is carried out on each irregular area, and edge pixel points in each irregular area are obtained; acquiring a target degree value of each irregular area according to the gray average value of each irregular area and the edge pixel points;
performing traversal merging on the irregular area, including:
s1: taking an irregular area with the maximum target degree value as a starting area; taking the initial region as the initial merging region;
s2: respectively taking all irregular areas adjacent to the initial area as areas to be selected; calculating a combined probability value between each region to be selected and the initial region; combining the to-be-selected region with the combination probability value larger than the preset combination threshold value with the combination region to which the initial region belongs; taking the region to be selected with the combination probability value larger than the preset combination threshold value as a new initial region;
s3: repeating S2 according to the new initial region, and stopping iteration until all irregular regions are not merged any more; taking the latest obtained combined area as a suspected hematoma area;
taking any outline pixel point of the suspected hematoma area as a pixel point to be measured, and carrying out outline adjustment on the pixel point to be measured, wherein the method comprises the following steps: acquiring a reference pixel point of a pixel point to be detected; acquiring all public pixel points according to the reference pixel points and the pixel points to be detected, and updating the suspected hematoma area according to all public pixel points and the pixel points to be detected;
and respectively taking each contour pixel point on the contour of the suspected hematoma area as a pixel point to be measured according to the clockwise direction, carrying out contour adjustment on each pixel point to be measured, and taking the finally obtained suspected hematoma area as an intracranial hematoma area to realize the segmentation of the intracranial hematoma area.
2. The method for segmenting intracranial hematoma areas based on CT images according to claim 1, wherein the step of obtaining the target degree value of each irregular area according to the gray average value of each irregular area and the edge pixel point comprises the following specific steps:
wherein a is k A target degree value for the kth irregular area; p is p k The gray average value of the kth irregular area; p is p max The maximum value of the gray average value of all the irregular areas; p is p ′ k，i A gray average value of an o-th irregular area adjacent to the k-th irregular area; n is n k The number of all irregular areas adjacent to the kth irregular area; c k The number of edge pixel points in the kth irregular area; d, d k The average value of Euclidean distances between every two edge pixel points in the kth irregular area; exp is an exponential function that bases on a natural constant.
3. The method for segmenting intracranial hematoma areas based on CT image according to claim 1, wherein the calculating of the combined probability value between each candidate area and the initial area comprises the following steps:
wherein sigma j A combined probability value between the j-th candidate region and the initial region; a, a ′ j The target degree value of the j-th candidate area is the target degree value; a is a target degree value of an initial region; p is p max The maximum value of the gray average value of all the irregular areas;the average value of the gray average value of the jth area to be selected and the gray average value of the initial area; is an absolute value symbol; exp is an exponential function that bases on a natural constant.
4. The method for segmenting an intracranial hematoma area based on a CT image according to claim 1, wherein the step of obtaining the reference pixel point of the pixel point to be detected comprises the following specific steps:
and acquiring two contour pixel points which are adjacent to the pixel point to be detected and are closest to the pixel point to be detected, and respectively taking the two contour pixel points as reference pixel points of the pixel point to be detected.
5. The method for segmenting an intracranial hematoma area based on a CT image according to claim 1, wherein the step of obtaining all the common pixels according to the reference pixel and the pixel to be detected comprises the following specific steps:
and taking the pixel points which are simultaneously positioned in eight adjacent areas of the pixel points to be detected and the two reference pixel points as the common pixel point of the pixel points to be detected.
6. The method for segmenting an intracranial hematoma area based on a CT image according to claim 1, wherein the updating the suspected hematoma area according to all the common pixels and the pixels to be detected comprises the following specific steps:
acquiring the pixel points to be detected and the gradient amplitude of each public pixel point, and if the gradient amplitude of the pixel points to be detected is greater than or equal to the gradient amplitude of each public pixel point, not updating the suspected hematoma area;
if the gradient amplitude of the pixel to be detected is smaller than the gradient amplitude of one or more public pixel points, the public pixel point with the largest gradient amplitude is obtained, when the public pixel point with the largest gradient amplitude does not belong to the suspected hematoma area, the public pixel point with the largest gradient amplitude is merged into the suspected hematoma area, and when the public pixel point with the largest gradient amplitude belongs to the suspected hematoma area, the pixel point to be detected is removed from the suspected hematoma area.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310460604.3A CN116485814A (en) | 2023-04-25 | 2023-04-25 | Intracranial hematoma region segmentation method based on CT image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310460604.3A CN116485814A (en) | 2023-04-25 | 2023-04-25 | Intracranial hematoma region segmentation method based on CT image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116485814A true CN116485814A (en) | 2023-07-25 |
Family
ID=87213479
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310460604.3A Pending CN116485814A (en) | 2023-04-25 | 2023-04-25 | Intracranial hematoma region segmentation method based on CT image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116485814A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117237389A (en) * | 2023-11-14 | 2023-12-15 | 深圳市亿康医疗技术有限公司 | CT image segmentation method for middle ear cholesteatoma |
| CN117557568A (en) * | 2024-01-12 | 2024-02-13 | 吉林省迈达医疗器械股份有限公司 | Focal region segmentation method in thermal therapy process based on infrared image |
| CN117911716A (en) * | 2024-03-19 | 2024-04-19 | 天津医科大学总医院 | Arthritis CT image feature extraction method based on machine vision |
-
2023
- 2023-04-25 CN CN202310460604.3A patent/CN116485814A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117237389A (en) * | 2023-11-14 | 2023-12-15 | 深圳市亿康医疗技术有限公司 | CT image segmentation method for middle ear cholesteatoma |
| 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 |
| CN117557568A (en) * | 2024-01-12 | 2024-02-13 | 吉林省迈达医疗器械股份有限公司 | Focal region segmentation method in thermal therapy process based on infrared image |
| CN117557568B (en) * | 2024-01-12 | 2024-05-03 | 吉林省迈达医疗器械股份有限公司 | Focal region segmentation method in thermal therapy process based on infrared image |
| CN117911716A (en) * | 2024-03-19 | 2024-04-19 | 天津医科大学总医院 | Arthritis CT image feature extraction method based on machine vision |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116485814A (en) | Intracranial hematoma region segmentation method based on CT image | |
| WO2021129323A1 (en) | Ultrasound image lesion describing method and apparatus, computer device, and storage medium | |
| US7689016B2 (en) | Automatic detection of critical dermoscopy features for malignant melanoma diagnosis | |
| CN116109663B (en) | Stomach CT image segmentation method based on multi-threshold segmentation | |
| CN115330800A (en) | Automatic segmentation method for radiotherapy target area based on image processing | |
| CN109816655B (en) | Pulmonary nodule image feature detection method based on CT image | |
| CN108615239B (en) | Tongue image segmentation method based on threshold technology and gray level projection | |
| JP2006517663A (en) | Image analysis | |
| US9898824B2 (en) | Method for volume evaluation of penumbra mismatch in acute ischemic stroke and system therefor | |
| US5960102A (en) | X-ray image processing method and device for performing that method in which a portion corresponding to an x-ray absorption filter is selectively processed | |
| CN106096491B (en) | Automatic identification method for microaneurysms in fundus color photographic image | |
| CN116269467A (en) | Information acquisition system before debridement of wounded patient | |
| CN116071355A (en) | Auxiliary segmentation system and method for peripheral blood vessel image | |
| CN116912255B (en) | Follicular region segmentation method for ovarian tissue analysis | |
| CN116152505A (en) | Bone target identification and segmentation method based on X-ray data | |
| KR101251822B1 (en) | System and method for analysising perfusion in dynamic contrast-enhanced lung computed tomography images | |
| CN108629780B (en) | Tongue image segmentation method based on color decomposition and threshold technology | |
| CN109447948B (en) | Optic disk segmentation method based on focus color retina fundus image | |
| CN113689424B (en) | Ultrasonic inspection system capable of automatically identifying image features and identification method | |
| CN112258533B (en) | Method for segmenting cerebellum earthworm part in ultrasonic image | |
| KR102380560B1 (en) | Corneal Ulcer Region Detection Apparatus Using Image Processing and Method Thereof | |
| CN112634280B (en) | MRI image brain tumor segmentation method based on energy functional | |
| Alves et al. | The challenges of applying deep learning for hemangioma lesion segmentation | |
| CN114155205A (en) | Mammary nodule boundary definition judging device | |
| CN112085711A (en) | Method for automatically tracking muscle pinnate angle by combining convolutional neural network and Kalman filtering |
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 |