CN110648338B - Image segmentation method, readable storage medium, and image processing apparatus - Google Patents
Image segmentation method, readable storage medium, and image processing apparatus Download PDFInfo
- Publication number
- CN110648338B CN110648338B CN201910913757.2A CN201910913757A CN110648338B CN 110648338 B CN110648338 B CN 110648338B CN 201910913757 A CN201910913757 A CN 201910913757A CN 110648338 B CN110648338 B CN 110648338B
- Authority
- CN
- China
- Prior art keywords
- image
- tissue organ
- region
- tissue
- seed 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.)
- Active
Links
Images
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
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—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
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10088—Magnetic resonance imaging [MRI]
-
- 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
- G06T2207/30008—Bone
-
- 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
- G06T2207/30048—Heart; Cardiac
-
- 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
- G06T2207/30056—Liver; Hepatic
-
- 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
- G06T2207/30061—Lung
-
- 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
- G06T2207/30084—Kidney; Renal
-
- 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
- G06T2207/30101—Blood vessel; Artery; Vein; Vascular
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 provides an image segmentation method, a readable storage medium and an image processing device.A first seed point is set in a boundary region of a first image, and a second tissue organ is distinguished by taking the first seed point as an initial point through a morphological method to obtain a second image; and then carrying out maximum connected domain analysis on the second image, taking the central point coordinate of the image subjected to the maximum connected domain analysis as a second seed point, taking the second seed point as an initial point of region growth, and obtaining an output image of the first tissue organ through iterative combination. In the whole image segmentation process, the first seed point and the second seed point can be selected according to the characteristic positions of the image, so that the seed points can be automatically selected, manual input of seed point information is not needed, time cost can be greatly saved, the workload of doctors is reduced, the man-machine interaction operation flow is reduced, and the diagnosis efficiency and accuracy are improved.
Description
Technical Field
The present invention relates to the field of image segmentation technologies, and in particular, to an image segmentation method, a readable storage medium, and an image processing apparatus.
Background
Vascular diseases, especially cardiovascular diseases, have become one of the major diseases threatening human life safety. During surgery, physicians are assisted in diagnosing various vascular diseases, such as calcification, aortic dissection, aneurysms, etc., by vascular imaging techniques. The blood vessel imaging techniques mainly include Computed Tomography Angiography (CTA), magnetic Resonance Angiography (MRA), and the like. Generally, a three-dimensional image is obtained by imaging a blood vessel, and not only blood vessel tissues but also other tissues (bones, fat, muscles, lung tissues and the like) around the blood vessel cannot bring accurate diagnosis to doctors. Therefore, the whole blood vessel region is extracted from the three-dimensional image, and the shape of the blood vessel is displayed by the three-dimensional display technology, so that the diagnosis accuracy of doctors can be improved.
Although there are many techniques for vessel segmentation, the vessel segmentation problem remains a very challenging task. At present, the blood vessel segmentation method mainly comprises manual operation and semi-automatic operation, and the existing semi-automatic blood vessel segmentation methods can be roughly divided into two types: top down and bottom up. In the prior art, a top-down semi-automatic segmentation method needs to manually input seed points as starting conditions, then iteratively merges adjacent regions based on target errors, and finally generates an image.
Disclosure of Invention
The invention aims to provide an image segmentation method, a readable storage medium and an image processing device, which aim to solve the problems of complicated operation and low efficiency of the conventional image segmentation method.
In order to solve the above-mentioned technical problem, the present invention provides an image segmentation method for segmenting a predetermined first tissue organ from an image, the first tissue organ being connected to a second tissue organ and a third tissue organ, respectively, the image segmentation method comprising:
segmenting a preprocessed first image according to a first threshold value to obtain a second image comprising the second tissue organ;
setting a first seed point in a boundary region of the second image, and morphologically distinguishing the second tissue organ by taking the first seed point as an initial point to obtain a third image containing the first tissue organ and the third tissue organ;
performing maximum connected domain analysis on the third image to take the central point coordinate of the image subjected to the maximum connected domain analysis on the third image as a second seed point;
and taking the second seed point as an initial point of region growing, and obtaining an output image of the first tissue organ through iterative combination.
Optionally, the process of preprocessing the first image includes:
and filtering noise information in the original image by adopting a three-dimensional Gaussian filter.
Optionally, the raw image comprises a CTA volume data image comprising an image of a human tissue region and an image of a CT bed region, wherein the human tissue region comprises the first, second and third tissue organs; the process of pre-processing the first image further comprises:
segmenting the image with the noise information filtered by the three-dimensional Gaussian filter according to a second threshold value to obtain a fourth image including the human tissue region;
and setting a third seed point in the boundary region of the fourth image, and taking the third seed point as an initial point to distinguish the CT bed region and the human tissue region by a morphological method.
Optionally, the method for distinguishing the CT bed region from the human tissue region by a morphological method includes a morphological connected domain operation and/or a morphological opening operation.
Optionally, the morphological opening operation comprises:
carrying out corrosion treatment on the fourth image;
and performing expansion processing on the fourth image subjected to the corrosion processing.
Optionally, after the CT bed region and the human tissue region are distinguished by a morphological method, the process of preprocessing the first image further includes: and adopting a morphological maximum connected domain operation to obtain an image of the human tissue region, wherein the image of the human tissue region is defined as the first image.
Optionally, the step of morphologically differentiating the second tissue organ comprises:
and distinguishing the area of the second tissue organ by taking the first seed point as an initial point and by a morphological water-gold mountain diffusion method.
Optionally, the second tissue organ includes two parts disposed at an interval, and the first tissue organ and the third tissue organ are both located between the two parts of the second tissue organ, and after distinguishing the region of the second tissue organ, the step of morphologically distinguishing the second tissue organ further includes:
a morphological closing operation is applied, based on a preset morphological parameter, to obtain a fifth image of said second tissue organ and of the connection zone between the two portions thereof.
Optionally, after acquiring the fifth image, the step of morphologically distinguishing the second tissue organ further includes:
using a binary mask to isolate said second tissue organ from said fifth image;
and removing the image of the second tissue organ in the fifth image after the binary mask is used by using a third threshold value to obtain a third image.
In order to solve the above technical problem, the present invention further provides a readable storage medium, on which a computer program is stored, which when executed, can implement the image segmentation method as described above.
In order to solve the above technical problem, the present invention further provides an image processing apparatus, which includes a processor and a memory, wherein the memory stores a computer program, and the computer program is executed by the processor to execute the image segmentation method.
In summary, in the image segmentation method, the readable storage medium and the image processing apparatus provided by the present invention, a first seed point is set in a boundary region of a first image, and a second tissue is morphologically distinguished by using the first seed point as an initial point to obtain a second image; and then carrying out maximum connected domain analysis on the second image, taking the central point coordinate of the image subjected to the maximum connected domain analysis as a second seed point, taking the second seed point as an initial point of region growth, and obtaining an output image of the first tissue organ through iterative combination. In the whole image segmentation process, the first seed point and the second seed point can be selected according to the characteristic positions of the image, so that the seed points can be automatically selected, manual input of seed point information is not needed, time cost can be greatly saved, the workload of doctors is reduced, the man-machine interaction operation flow is reduced, and the diagnosis efficiency and accuracy are improved.
Drawings
It will be appreciated by those skilled in the art that the drawings are provided for a better understanding of the invention and do not constitute any limitation to the scope of the invention. Wherein:
fig. 1 is a schematic diagram of an original image in an image segmentation method according to an embodiment of the present invention;
FIG. 2 is a diagram illustrating a first image in an image segmentation method according to an embodiment of the present invention;
FIG. 3 is a diagram illustrating a third image in an image segmentation method according to an embodiment of the present invention;
FIG. 4 is a diagram illustrating a second seed point in an image segmentation method according to an embodiment of the present invention;
FIG. 5 is an output image of a first tissue organ in an image segmentation method according to an embodiment of the present invention;
fig. 6 is a flowchart of an image segmentation method according to an embodiment of the present invention.
In the drawings:
100-human tissue region; 200-CT bed area; 300-a third tissue organ; 400-a second seed point; 500-first tissue organ.
Detailed Description
To further clarify the objects, advantages and features of the present invention, a more particular description of the invention will be rendered by reference to specific embodiments thereof which are illustrated in the appended drawings. It is to be noted that the drawings are in simplified form and are not to scale, but are provided for the purpose of facilitating and clearly illustrating embodiments of the present invention. Further, the structures illustrated in the drawings are often part of actual structures. In particular, the drawings may have different emphasis points and may sometimes be scaled differently.
As used in this specification and the appended claims, the singular forms "a", "an", and "the" include plural referents unless the content clearly dictates otherwise. As used in this specification and the appended claims, the term "or" is generally employed in its sense including "and/or" unless the content clearly dictates otherwise.
The core idea of the present invention is to provide an image segmentation method, a readable storage medium and an image processing apparatus, so as to solve the problems of complicated operation and low efficiency of the conventional image segmentation method. The image segmentation method is used for segmenting a predetermined first tissue organ from an image, wherein the first tissue organ is respectively connected with a second tissue organ and a third tissue organ; the image segmentation method comprises the following steps: segmenting a preprocessed first image according to a first threshold value to obtain a second image comprising the second tissue organ; setting a first seed point in a boundary region of the second image, and morphologically distinguishing the second tissue organ by taking the first seed point as an initial point to obtain a third image containing the first tissue organ and the second tissue organ; performing maximum connected domain analysis on the third image to take the center point coordinate of the image subjected to the maximum connected domain analysis on the third image as a second seed point; and taking the second seed point as an initial point of region growing, and obtaining an output image of the first tissue organ through iterative combination. In the whole image segmentation process, the first seed point and the second seed point can be selected according to the characteristic positions of the image, so that the seed points can be automatically selected, manual input of seed point information is not needed, time cost can be greatly saved, the workload of doctors is reduced, the man-machine interaction operation flow is reduced, and the diagnosis efficiency and accuracy are improved.
The following description refers to the accompanying drawings.
Referring to fig. 1 to 6, wherein fig. 1 is a schematic diagram of an original image in an image segmentation method according to an embodiment of the present invention, fig. 2 is a schematic diagram of a first image in the image segmentation method according to the embodiment of the present invention, fig. 3 is a schematic diagram of a third image in the image segmentation method according to the embodiment of the present invention, fig. 4 is a schematic diagram of a second sub-point in the image segmentation method according to the embodiment of the present invention, fig. 5 is an output image of a first tissue in the image segmentation method according to the embodiment of the present invention, and fig. 6 is a flowchart of the image segmentation method according to the embodiment of the present invention.
As shown in fig. 6, an embodiment of the present invention provides an image segmentation method for segmenting a predetermined first tissue organ from an image, the first tissue organ being connected to a second tissue organ and a third tissue organ, respectively. The image segmentation method comprises the following steps:
step S1: segmenting a preprocessed first image according to a first threshold value to obtain a second image comprising a second tissue organ;
step S2: setting a first seed point in a boundary region of the second image, and morphologically distinguishing a second tissue organ by using the first seed point as an initial point to obtain a third image including the first tissue organ and the second tissue organ, as shown in fig. 3;
and step S3: performing maximum connected component analysis on the third image to use the coordinates of the center point of the image after performing the maximum connected component analysis on the third image as a second seed point, as shown in fig. 4;
and step S4: and taking the second seed point as an initial point of region growing, and obtaining an output image of the first tissue organ by iterative combination, as shown in fig. 5.
The following describes the image segmentation method provided in this embodiment by taking blood vessels as a first tissue organ, lungs as a second tissue organ, and heart as a third tissue organ as an illustrative example. It is to be understood that blood vessels used herein as the first tissue organ specifically refer to blood vessels in the cardiopulmonary area.
Step S1: first, a preprocessed first image is obtained, as shown in fig. 2, and the first image may be an image of a preprocessed human tissue region 100. The tissue region of the human body comprises at least a first tissue organ 500 (blood vessel), a second tissue organ (lung) and a third tissue organ 300 (heart), and typically the first image further comprises other tissue organs of the human body, such as bone, muscle, kidney, liver or fat, etc. Since the present example is mainly used for acquiring the output image of the first tissue organ 500 (blood vessel), the images of other tissue organs need to be segmented. Typically, the CT image is calibrated to represent different regions or tissues, for example, a pixel value of 0 represents water, and a pixel value of-600 to-400 represents lung tissue, so that the pixel value corresponding to lung tissue can be selected as the first threshold to segment the first image. In particular, the first threshold may be selected in accordance with the target segmented tissue, e.g. -500, i.e. the second tissue organ (lung) may be segmented from other tissues. Thereby, a second image including a second tissue organ (lung) is obtained.
Step S2: and setting a first seed point in the boundary area of the second image. In general, in the second image, the image of the second tissue organ (lung) is located in the central portion, and at this time, the first seed point is set in the boundary region of the second image, so that the first seed point can be ensured to be located outside the lung region, and the second tissue organ (lung) can be further distinguished by a morphological method, thereby obtaining a third image including the first tissue organ (blood vessel) and the third tissue organ (heart). It will be appreciated that this third image is a preliminary image containing the heart and blood vessels, as shown in fig. 3. The first seed point may be selected as a position of a specific pixel point in the third image, and the first seed point represents a region other than the second tissue organ (lung) and is used as an input parameter for subsequently distinguishing the second tissue organ (lung). The morphological method in step S2 may be a connected domain method, and the specific operation is to determine whether the pixel of the first seed point is the same as the adjacent pixel, and if so, the first seed point is considered to be a region of the second tissue organ (lung), otherwise, the first seed point is determined to be a region of the second tissue organ (lung); this determination is repeated iteratively throughout the third image to distinguish regions of the second tissue organ (lung). The first seed point in the step is not required to be selected manually, but the image boundary area is set as the seed point in advance, so that the method can greatly save time and cost and reduce the workload of doctors.
And step S3: on the basis of the third image obtained in step S2, i.e. the image of the preliminarily segmented heart vessels, a maximum connected component analysis is performed. During specific operation, the number of pixels in each connected domain (one connected domain represents a set with the same pixels) is counted, wherein the largest number of pixels is the largest connected domain. By performing a maximum connected component analysis on the third image, small target areas, such as bone areas close to the lungs, can be removed. Then, the two-dimensional data of the intermediate layer, the two-dimensional data of 1/3 layer, and the two-dimensional data of 2/3 layer of the volume data (the three-dimensional data of the third image obtained in step S2, for example, the three-dimensional image of the cardiac blood vessel) are selected, the maximum connected component analysis is performed, and the center point coordinates of the image after the maximum connected component analysis are used as the second seed points 400, as shown in fig. 4. In general, the coordinates of the center point of the maximum connected component image can be determined to be on the first tissue organ (blood vessel), and thus can be used as the second seed point 400 for region growing in the subsequent blood vessel image. In step S3, the acquisition of the second seed point is automatically calculated without manually inputting the seed point.
And step S4: region growing refers to the process of developing groups of pixels or regions into larger regions, i.e. starting from a set of seed points, from which the region grows by merging into this region neighboring pixels with similar properties like intensity, grey level, texture color etc. as each seed point. Therefore, the second seed point is used as the initial point of region growing, and the output image of the first tissue organ (blood vessel) can be obtained through iterative combination.
The above examples are examples of obtaining a blood vessel image from a CT image, but the present invention is not limited to this application. The invention can also be applied to image segmentation of any other tissue organ. For example, a first threshold value may be set to-900 (pixel values of the airway tube in the CT image) to segment the image of the airway tube.
Preferably, the process of preprocessing the first image comprises: and filtering noise information in the original image by adopting a three-dimensional Gaussian filter. Here, the original image mainly includes a CTA (computed tomography angiography) volume data (for example, three-dimensional data of an image) image, and the size is selected to be 512 × 130, for example, although the size may be selected according to a specific image in implementation. Further, referring to fig. 1, the CTA volume data image includes an image of a human tissue region 100 and an image of a CT bed region 200, wherein the human tissue region 100 includes the first tissue organ, the second tissue organ and the third tissue organ. For the convenience of subsequent processing analysis, the CT bed region 200 needs to be removed and only the image of the body region 100 is retained. The CT bed region 200 can be separated from the image of the body tissue region 100 by:
step Sa1: and segmenting the image with the noise information filtered by the three-dimensional Gaussian filter according to a second threshold value to obtain a fourth image including the human tissue region 100. Specifically, the selection of the second threshold may be performed according to a difference between a pixel value of the human tissue in the CT image and a pixel value of the CT bed in the CT image, so as to segment the image of the CT bed region 200, and thus, a fourth image including the human tissue region 100 may be obtained.
Step Sa2: and setting a third seed point in the boundary region of the fourth image, and taking the third seed point as an initial point to distinguish the CT bed region 200 and the human tissue region 100 by a morphological method. The morphological method adopted here may be similar to the connected domain method in step S2, and determine whether the pixel of the third sub-point is the same as the adjacent pixel, if so, it is considered as a non-human tissue region, otherwise, it is determined as a human tissue region; and continuously traversing and iterating the fourth image to repeat the judgment so as to distinguish the region of the human tissue. The third sub-point in this step also does not need to be selected manually, but the image boundary area is set as the seed point in advance.
Optionally, in the process of morphologically distinguishing the CT bed region from the human tissue region, the adopted morphology method further includes a morphology opening operation, sometimes the CT bed region 200 and the human tissue region 100 are connected together, and to solve this, the morphology opening operation is adopted to set the morphology parameters, for example, the morphology parameters are set to 25, so that the CT bed region 200 can be separated from the human tissue region 100. The common image morphological operations comprise corrosion, expansion, opening operation, closing operation and the like, wherein the morphological opening operation is adopted, namely, firstly, the corrosion processing is carried out on the image, and then, the expansion processing is carried out on the corroded image; the morphological parameters refer to the number of pixels, for example, when the fourth image is corroded, 25 pixels are corroded from the fourth image, so that the joint of the CT bed region and the human tissue region is completely disconnected; the fourth image after erosion is then expanded, for example by 25 pixels, to restore the tissue region to the pre-erosion state. In specific operation, for example, the image is divided into a plurality of layers from outside to inside, each layer represents 1 pixel point, the morphological parameter is set to 25, which means that 25 pixels are etched from outside to inside, and then the etched image is expanded by 25 pixels. Of course, in the process of distinguishing the CT bed region from the human tissue region by the morphological method, the adopted morphological method can also be used together with the connected domain operation and the morphological opening operation.
Further, after the CT bed region and the human tissue region are distinguished by a morphological method, the process of preprocessing the first image further includes: a morphological maximum connected component operation is used to obtain an image of the human tissue region, which is defined as the first image, as shown in fig. 2. The morphological maximum connected component operation here is similar to that in step S3, and therefore will not be described further.
Optionally, in step S2, the step of morphologically distinguishing the second tissue organ includes: and distinguishing the area of the second tissue organ by taking the first seed point as an initial point through a morphological water-diffusion hill method (FloodFill). Starting from a first sub-point of the boundary of the second image, obtaining an image with a lung region displayed as black and other regions displayed as white by a morphological water-flooding golden hill method, and performing an inversion operation on the obtained image (for example, performing a black-white image inversion operation to display the lung region as white and other regions displayed as black) to segment the lung region; the water-flooding golden hill method can be understood as that whether a first seed point is the same as an adjacent pixel or not is judged, if the first seed point is the same as the adjacent pixel, the first seed point is considered to be the same region, otherwise, the first seed point is a different region, then whether the region where the first seed point is located and the adjacent region have the same attribute (for example, the same pixel value) or not is continuously judged, and if the first seed point is different from the adjacent region, the region where the first seed point is located is set as the adjacent region attribute, so that the lung region and other regions are distinguished.
Further, the second tissue organ (lung) includes two parts (left lung and right lung) arranged at an interval, and the first tissue organ (blood vessel) and the third tissue organ (heart) are both located between the two parts of the second tissue organ, and after distinguishing the region of the second tissue organ, the step of morphologically distinguishing the second tissue organ further includes: a morphological closing operation is applied, based on a preset morphological parameter, to obtain a fifth image of said second tissue organ and of the connection zone between the two portions thereof. Generally, the heart blood vessels are between the left lung and the right lung of a human body, in order to keep the blood vessels, the morphological closing operation is adopted, and morphological parameters are set to be 25 for example, so that the areas between the two lungs can be connected together; specifically, in the morphological closing operation, an image of the lung region obtained by the morphological water-flooding-gold-hill method is expanded by, for example, 25 pixels, and then the expanded image is eroded by, for example, 25 pixels, to obtain a fifth image connecting the regions between the two lungs together.
Further, after acquiring the fifth image, the step of morphologically distinguishing the second tissue organ further comprises: masking said fifth image with a binary (non-black or white image, black and white) mask to isolate said second tissue organ; and then removing the image of the second tissue organ in the fifth image after the binary mask is used by using a third threshold value to obtain a third image. Here, the third threshold value may be chosen to be-500, since the object to be removed and isolated is mainly also the lung. Based on the above configuration, a third image (image of the cardiac blood vessel) can be obtained, as shown in fig. 3.
By the method, the seed points can be automatically selected, and the image can be automatically segmented. In practice, the method can be programmed into software, i.e. the image can be automatically segmented. Based on this, the present embodiment also provides a readable storage medium, on which a computer program is stored, which when executed, is capable of implementing the image segmentation method as described above. Furthermore, the software may be integrated into a hardware device, and thus, the embodiment further provides an image processing apparatus, which includes a processor and a memory, where the memory stores a computer program, and when the computer program is executed by the processor, the image segmentation method as described above is performed.
In summary, in the image segmentation method, the readable storage medium and the image processing apparatus provided by the present invention, a first seed point is set in a boundary region of a first image, and a second tissue is morphologically distinguished by using the first seed point as an initial point to obtain a second image; and then carrying out maximum connected domain analysis on the second image, taking the central point coordinate of the image after the maximum connected domain analysis as a second seed point, taking the second seed point as an initial point of region growth, and obtaining an output image of the first tissue organ through iterative combination. In the whole image segmentation process, the first seed point and the second seed point can be selected according to the characteristic positions of the image, so that the seed points can be automatically selected, manual input of seed point information is not needed, time cost can be greatly saved, the workload of doctors is reduced, the man-machine interaction operation flow is reduced, and the diagnosis efficiency and accuracy are improved.
The above description is only for the purpose of describing the preferred embodiments of the present invention, and is not intended to limit the scope of the present invention, and any variations and modifications made by those skilled in the art based on the above disclosure are within the scope of the appended claims.
Claims (9)
1. An image segmentation method for segmenting a predetermined first tissue organ from an image of a heart-lung region, the first tissue organ being connected to a second tissue organ and a third tissue organ, respectively, wherein the first tissue organ is a blood vessel, the second tissue organ is a lung, and the third tissue organ is a heart, the image segmentation method comprising:
segmenting a preprocessed first image according to a first threshold value to obtain a second image comprising the second tissue organ;
setting a first seed point in a boundary region of the second image, taking the first seed point as an initial point, and distinguishing the second tissue organ by a morphological method to obtain a third image containing the first tissue organ and the third tissue organ;
performing maximum connected domain analysis on the third image to take the center point coordinate of the image subjected to the maximum connected domain analysis on the third image as a second seed point;
taking the second seed point as an initial point of region growth, and obtaining an output image of the first tissue organ through iterative combination;
the process of preprocessing the first image comprises the steps of filtering noise information in an original image by adopting a three-dimensional Gaussian filter;
the raw images comprise CTA volumetric data images comprising an image of a human tissue region and an image of a CT bed region, wherein the human tissue region comprises the first, second, and third tissue organs; the process of pre-processing the first image further comprises:
segmenting the image with the noise information filtered by the three-dimensional Gaussian filter according to a second threshold value to obtain a fourth image including the human tissue region;
and setting a third seed point in the boundary region of the fourth image, and taking the third seed point as an initial point to distinguish the CT bed region and the human tissue region by a morphological method.
2. The image segmentation method according to claim 1, wherein the method for morphologically distinguishing the CT bed region from the human tissue region includes a morphological connected domain operation and/or a morphological open operation.
3. The image segmentation method according to claim 2, characterized in that the morphological opening operation comprises:
performing corrosion treatment on the fourth image;
and performing expansion processing on the fourth image subjected to the corrosion processing.
4. The image segmentation method according to claim 1, wherein after the CT bed region is morphologically distinguished from the human tissue region, the preprocessing the first image further comprises: and adopting a morphological maximum connected domain operation to obtain an image of the human tissue region, wherein the image of the human tissue region is defined as the first image.
5. The image segmentation method according to claim 1, wherein the step of morphologically distinguishing the second tissue organ comprises:
and distinguishing the area of the second tissue organ by taking the first seed point as an initial point and by a morphological water-gold mountain diffusion method.
6. The image segmentation method according to claim 5, wherein the second tissue organ includes two portions spaced apart from each other, and the first tissue organ and the third tissue organ are located between the two portions of the second tissue organ, and wherein the step of morphologically distinguishing the second tissue organ further includes:
a morphological closing operation is applied, based on a preset morphological parameter, to obtain a fifth image of said second tissue organ and of the connection zone between the two portions thereof.
7. The image segmentation method according to claim 6, wherein the step of morphologically distinguishing the second tissue organ after acquiring the fifth image further comprises:
using a binary mask to isolate said second tissue organ from said fifth image;
and removing the image of the second tissue organ in the fifth image after the binary mask is used by using a third threshold value to obtain a third image.
8. A readable storage medium on which a computer program is stored, the computer program being capable of implementing the image segmentation method according to any one of claims 1 to 7 when executed.
9. An image processing apparatus comprising a processor and a memory, the memory having stored thereon a computer program which, when executed by the processor, performs the image segmentation method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910913757.2A CN110648338B (en) | 2019-09-25 | 2019-09-25 | Image segmentation method, readable storage medium, and image processing apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910913757.2A CN110648338B (en) | 2019-09-25 | 2019-09-25 | Image segmentation method, readable storage medium, and image processing apparatus |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110648338A CN110648338A (en) | 2020-01-03 |
| CN110648338B true CN110648338B (en) | 2023-02-17 |
Family
ID=69011372
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910913757.2A Active CN110648338B (en) | 2019-09-25 | 2019-09-25 | Image segmentation method, readable storage medium, and image processing apparatus |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110648338B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113724264B (en) * | 2020-05-26 | 2023-08-11 | 上海微创卜算子医疗科技有限公司 | Image segmentation method, system and readable storage medium |
| CN112767332B (en) * | 2021-01-08 | 2024-05-17 | 心医国际数字医疗系统(大连)有限公司 | Blood vessel region judging method and system based on CTA image |
| CN112884792B (en) * | 2021-02-02 | 2022-10-25 | 青岛海信医疗设备股份有限公司 | Lung image segmentation method and device, electronic equipment and storage medium |
| CN113592819A (en) * | 2021-07-30 | 2021-11-02 | 上海皓桦科技股份有限公司 | Image processing system and method |
| CN113744849B (en) * | 2021-08-31 | 2024-03-19 | 华中科技大学 | Semi-automatic labeling method and system for CT image |
| CN114299081B (en) * | 2021-12-16 | 2023-02-17 | 北京朗视仪器股份有限公司 | Maxillary sinus CBCT image segmentation method, maxillary sinus CBCT image segmentation device, maxillary sinus CBCT storage medium and electronic equipment |
| CN114119602B (en) * | 2021-12-20 | 2022-04-15 | 深圳科亚医疗科技有限公司 | Method, apparatus and storage medium for object analysis of medical images |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104143035A (en) * | 2013-05-10 | 2014-11-12 | 上海联影医疗科技有限公司 | Method for partitioning breast lesion |
| CN106780460A (en) * | 2016-12-13 | 2017-05-31 | 杭州健培科技有限公司 | A kind of Lung neoplasm automatic checkout system for chest CT image |
| CN107481251A (en) * | 2017-07-17 | 2017-12-15 | 东北大学 | A kind of method that terminal bronchi tree is extracted from lung CT image |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104992445B (en) * | 2015-07-20 | 2017-10-20 | 河北大学 | A kind of automatic division method of CT images pulmonary parenchyma |
-
2019
- 2019-09-25 CN CN201910913757.2A patent/CN110648338B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104143035A (en) * | 2013-05-10 | 2014-11-12 | 上海联影医疗科技有限公司 | Method for partitioning breast lesion |
| CN106780460A (en) * | 2016-12-13 | 2017-05-31 | 杭州健培科技有限公司 | A kind of Lung neoplasm automatic checkout system for chest CT image |
| CN107481251A (en) * | 2017-07-17 | 2017-12-15 | 东北大学 | A kind of method that terminal bronchi tree is extracted from lung CT image |
Non-Patent Citations (3)
| Title |
|---|
| 基于三维区域生长法的 CT 图像肺部血管分割;冯宗雪;《中国优秀博硕士学位论文全文数据库（硕士） 医药卫生科技辑》;20180115;第13-24页 * |
| 基于区域生长的肝脏CT图像肿瘤分割方法研究;郭颖;《中国优秀硕士学位论文全文数据库》;20170715;第24-34页 * |
| 基于漫水填充算法的肺实质分割方法;冯宗雪等;《电脑知识与技术》;20160630;第205-207页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110648338A (en) | 2020-01-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110648338B (en) | Image segmentation method, readable storage medium, and image processing apparatus | |
| Rusko et al. | Fully automatic liver segmentation for contrast-enhanced CT images | |
| CN110706246B (en) | Blood vessel image segmentation method and device, electronic equipment and storage medium | |
| Bauer et al. | Segmentation of interwoven 3d tubular tree structures utilizing shape priors and graph cuts | |
| US8761475B2 (en) | System and method for automatic recognition and labeling of anatomical structures and vessels in medical imaging scans | |
| Yi et al. | A locally adaptive region growing algorithm for vascular segmentation | |
| Manniesing et al. | Vessel axis tracking using topology constrained surface evolution | |
| JP6570145B2 (en) | Method, program, and method and apparatus for constructing alternative projections for processing images | |
| US20060013482A1 (en) | System and methods of organ segmentation and applications of same | |
| EP3497669B1 (en) | Method for automatically detecting systemic arteries in arbitrary field-of-view computed tomography angiography (cta). | |
| WO2022105623A1 (en) | Intracranial vascular focus recognition method based on transfer learning | |
| US11830193B2 (en) | Recognition method of intracranial vascular lesions based on transfer learning | |
| CN111815663A (en) | Hepatic vessel segmentation system based on Hessian matrix and gray scale method | |
| Vukadinovic et al. | Segmentation of the outer vessel wall of the common carotid artery in CTA | |
| CN114170151A (en) | Intracranial vascular lesion identification method based on transfer learning | |
| CN112562058B (en) | Method for quickly establishing intracranial vascular simulation three-dimensional model based on transfer learning | |
| Boskamp et al. | Geometrical and structural analysis of vessel systems in 3D medical image datasets | |
| CN115775219A (en) | Medical image segmentation method, system, electronic device, and medium | |
| CN114170337A (en) | Method for establishing intracranial vascular enhancement three-dimensional model based on transfer learning | |
| CN117078711A (en) | Medical image segmentation method, system, electronic device and storage medium | |
| CN112634386A (en) | Method for establishing angiography enhanced three-dimensional narrowing analysis model | |
| Orkisz et al. | Advances of blood vessel morphology analysis in 3D magnetic resonance images | |
| CN112509080A (en) | Method for establishing intracranial vascular simulation three-dimensional model based on transfer learning | |
| Descoteaux | A multi-scale geometric flow for segmenting vasculature in mri: Theory and validation | |
| Luo et al. | Recent progresses on cerebral vasculature segmentation for 3D quantification and visualization of MRA |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20210812 Address after: 201203 room 1702, building 1, No. 1601, Zhangdong Road, China (Shanghai) pilot Free Trade Zone, Pudong New Area, Shanghai (actual floor 15) Applicant after: Shanghai Weiwei Medical Technology Co.,Ltd. Address before: 201203 No. 1601 Zhangdong Road, Zhangjiang High-tech Park, Pudong New Area, Shanghai Applicant before: SHANGHAI MICROPORT MEDICAL (Group) Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |