CN114972257A - Coronary artery segmentation method, coronary artery segmentation device, storage medium, and electronic apparatus - Google Patents

Coronary artery segmentation method, coronary artery segmentation device, storage medium, and electronic apparatus Download PDF

Info

Publication number
CN114972257A
CN114972257A CN202210585956.7A CN202210585956A CN114972257A CN 114972257 A CN114972257 A CN 114972257A CN 202210585956 A CN202210585956 A CN 202210585956A CN 114972257 A CN114972257 A CN 114972257A
Authority
CN
China
Prior art keywords
image
domain
coronary artery
lung
coronary
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
Application number
CN202210585956.7A
Other languages
Chinese (zh)
Inventor
平安
彭成宝
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenyang Neusoft Intelligent Medical Technology Research Institute Co Ltd
Original Assignee
Shenyang Neusoft Intelligent Medical Technology Research Institute Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shenyang Neusoft Intelligent Medical Technology Research Institute Co Ltd filed Critical Shenyang Neusoft Intelligent Medical Technology Research Institute Co Ltd
Priority to CN202210585956.7A priority Critical patent/CN114972257A/en
Publication of CN114972257A publication Critical patent/CN114972257A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • 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
    • G06T5/00—Image enhancement or restoration
    • 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
    • G06T2207/30—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
    • G06T2207/30061—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
    • G06T2207/30101—Blood vessel; Artery; Vein; Vascular

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Radiology & Medical Imaging (AREA)
  • Quality & Reliability (AREA)
  • Apparatus For Radiation Diagnosis (AREA)
  • Image Processing (AREA)

Abstract

The present disclosure relates to a coronary artery segmentation method, apparatus, storage medium, and electronic device, the method comprising: acquiring an original image containing a cardiovascular region; inputting the original image into a pre-trained coronary artery prediction model to obtain a coronary artery prediction image output by the coronary artery prediction model, wherein the coronary artery prediction image is used for predicting coronary arteries in the cardiovascular region; carrying out image enhancement processing on the heart blood vessel region in the original image to obtain a coronary artery enhanced image; and according to the coronary artery prediction image and the coronary artery enhanced image, carrying out segmentation processing on the original image to obtain a coronary artery segmentation image corresponding to the original image. In this way, the coronary artery in the coronary artery prediction image is restored by the coronary artery enhanced image, and a more accurate coronary artery segmentation result can be obtained, thereby improving the effect of coronary artery segmentation.

Description

Coronary artery segmentation method, coronary artery segmentation device, storage medium, and electronic apparatus
Technical Field
The present disclosure relates to the field of medical imaging technologies, and in particular, to a coronary artery segmentation method, apparatus, storage medium, and electronic device.
Background
Image processing technology is mostly adopted in early stage of coronary artery segmentation, and coronary arteries are extracted based on coronary artery vessel characteristics, but most of the coronary arteries are interactive, and accuracy of automatic extraction is difficult to guarantee. With the development of deep learning technology, coronary artery segmentation based on the deep learning technology is widely applied.
However, since the coronary artery is thin, the end thin vessel is difficult to label, no labeled data is available, and the model cannot be learned, the coronary artery segmentation method based on the deep learning technique can only segment the main coronary artery and the branch vessel accurately, and the effect of coronary artery segmentation is poor.
Disclosure of Invention
In order to solve the above problems, the present disclosure provides a coronary artery segmentation method, apparatus, storage medium, and electronic device.
In a first aspect, the present disclosure provides a coronary artery segmentation method, comprising:
acquiring an original image containing a cardiovascular region;
inputting the original image into a pre-trained coronary artery prediction model to obtain a coronary artery prediction image output by the coronary artery prediction model, wherein the coronary artery prediction image is used for predicting coronary arteries in the cardiovascular region;
carrying out image enhancement processing on the heart blood vessel region in the original image to obtain a coronary artery enhanced image;
and according to the coronary artery prediction image and the coronary artery enhanced image, carrying out segmentation processing on the original image to obtain a coronary artery segmentation image corresponding to the original image.
Optionally, the performing image enhancement processing on the cardiovascular region in the original image to obtain a coronary artery enhanced image includes:
determining a blood vessel domain enhanced image containing the heart blood vessel region according to the original image;
and determining the coronary artery enhanced image according to preset position information and the blood vessel domain enhanced image, wherein the preset position information is preset position information of a coronary artery region in the original image.
Optionally, the determining, from the original image, a vessel domain enhanced image containing the cardiovascular region comprises:
determining a heart domain image containing a heart region from the original image;
enhancing the heart domain image to obtain a heart domain enhanced image;
carrying out segmentation processing on the heart domain enhanced image to obtain a tubular domain enhanced image containing a tubular region;
and determining the blood vessel domain enhanced image according to a preset blood vessel parameter threshold value and the tubular domain enhanced image.
Optionally, the determining, from the original image, a heart domain image containing a heart region comprises:
determining a lung area image containing a lung area according to the original image;
and extracting the heart domain image from the lung domain image according to preset heart position information.
Optionally, the determining, according to the original image, a lung region image including a lung region includes:
determining a lung area binary image corresponding to the original image according to a preset lung area parameter threshold;
determining a plurality of lung connected domain images in the lung domain binary image;
and determining the lung domain image according to the lung connected domain images.
Optionally, the determining the lung domain image according to the plurality of lung connected domain images comprises:
determining a first length value corresponding to a first lung connected domain image, wherein the first lung connected domain image is the largest lung connected domain image in the plurality of lung connected domain images, and the first length value is the length of the first lung connected domain image in a first direction;
and taking the first lung connected domain image as the lung domain image when the difference between the first length value and a second length value is greater than or equal to a preset difference threshold, wherein the second length value is the length of the original image in the first direction.
Optionally, the method further comprises:
and under the condition that the difference value between the first length value and the second length value is smaller than the preset difference threshold value, splicing the first lung connected domain image and the second lung connected domain image to obtain the lung domain image, wherein the second lung connected domain image is the largest lung connected domain image except the first lung connected domain image in the plurality of lung connected domain images.
Optionally, before the extracting the heart domain image from the lung domain image according to the preset heart position information, the method further includes:
filling holes in the lung area image to obtain a target lung area image;
the extracting the heart domain image from the lung domain image according to the preset heart position information comprises:
and extracting the heart domain image from the target lung domain image according to the heart position information.
Optionally, the determining the vessel domain enhanced image according to the preset vessel parameter threshold and the tubular domain enhanced image includes:
and taking an image formed by pixel points of which the parameter values of the preset parameters in the tubular domain enhanced image are greater than or equal to the preset blood vessel parameter threshold value as the blood vessel domain enhanced image.
Optionally, the determining the coronary artery enhanced image according to the preset position information and the vessel domain enhanced image includes:
determining a plurality of vessel domain connected images corresponding to the vessel domain enhanced image;
determining a plurality of target blood vessel domain connected images from the blood vessel domain connected images according to the preset position information;
and taking the union of a plurality of target blood vessel domain connected images as the coronary enhanced image.
Optionally, the segmenting the original image according to the predicted coronary artery image and the enhanced coronary artery image to obtain a segmented coronary artery image corresponding to the original image includes:
determining a main coronary artery image corresponding to the coronary artery prediction image;
determining a plurality of coronary vessel domain connected images corresponding to the coronary artery enhanced image;
and according to the coronary artery main branch image and the coronary artery vessel domain communication images, carrying out segmentation processing on the original image to obtain the coronary artery segmentation image.
Optionally, the segmenting the original image according to the main coronary artery map and the multiple coronary vessel domain connected images to obtain the coronary artery segmentation image includes:
taking the largest coronary vessel domain connected image in the plurality of coronary vessel domain connected images as a target connected image, and circularly executing an image processing step according to the target connected image and the coronary artery main branch image until each coronary vessel domain connected image executes the image processing step, and taking a new coronary artery main branch image as the coronary artery segmentation image;
wherein the image processing step comprises:
determining an intersection image of the target connected image and the main coronary artery image;
determining whether a connected domain exists in the intersection image;
taking the union of the target connected image and the main coronary artery image as a new main coronary artery image and the largest connected coronary artery image in the rest connected coronary artery images as a new target connected image under the condition that the intersection images have connected domains, wherein the rest connected coronary artery images comprise the connected coronary artery images which are not subjected to the image processing step in the plurality of connected coronary artery images;
and under the condition that the intersection images do not have the connected image, taking the largest coronary vessel region connected image in the residual coronary vessel region connected images as a new target connected image.
In a second aspect, the present disclosure provides a coronary artery segmentation apparatus comprising:
a first acquisition module for acquiring an original image containing a cardiovascular region;
a second obtaining module, configured to input the original image into a pre-trained coronary artery prediction model to obtain a coronary artery prediction image output by the coronary artery prediction model, where the coronary artery prediction image is used to predict coronary arteries in the cardiovascular region;
the enhancement processing module is used for carrying out image enhancement processing on the heart blood vessel region in the original image to obtain a coronary artery enhancement image;
and the segmentation module is used for carrying out segmentation processing on the original image according to the coronary predicted image and the coronary enhanced image to obtain a coronary segmentation image corresponding to the original image.
Optionally, the enhancement processing module is further configured to:
determining a blood vessel domain enhanced image containing the heart blood vessel region according to the original image;
and determining the coronary artery enhanced image according to preset position information and the blood vessel domain enhanced image, wherein the preset position information is preset position information of a coronary artery region in the original image.
Optionally, the enhancement processing module is further configured to:
determining a heart domain image containing a heart region from the original image;
enhancing the heart domain image to obtain a heart domain enhanced image;
carrying out segmentation processing on the heart domain enhanced image to obtain a tubular domain enhanced image containing a tubular region;
and determining the blood vessel domain enhanced image according to a preset blood vessel parameter threshold value and the tubular domain enhanced image.
Optionally, the enhancement processing module is further configured to:
determining a lung area image containing a lung area according to the original image;
and extracting the heart domain image from the lung domain image according to preset heart position information.
Optionally, the enhancement processing module is further configured to:
determining a lung area binary image corresponding to the original image according to a preset lung area parameter threshold;
determining a plurality of lung connected domain images in the lung domain binary image;
and determining the lung domain image according to the lung connected domain images.
Optionally, the enhancement processing module is further configured to:
determining a first length value corresponding to a first lung connected domain image, wherein the first lung connected domain image is the largest lung connected domain image in the plurality of lung connected domain images, and the first length value is the length of the first lung connected domain image in a first direction;
and taking the first lung connected domain image as the lung domain image when the difference between the first length value and a second length value is greater than or equal to a preset difference threshold, wherein the second length value is the length of the original image in the first direction.
Optionally, the enhancement processing module is further configured to:
and under the condition that the difference value between the first length value and the second length value is smaller than the preset difference threshold value, splicing the first lung connected domain image and the second lung connected domain image to obtain the lung domain image, wherein the second lung connected domain image is the largest lung connected domain image except the first lung connected domain image in the plurality of lung connected domain images.
Optionally, the apparatus further comprises:
the filling module is used for filling holes in the lung domain image to obtain a target lung domain image;
the enhancement processing module is further configured to:
and extracting the heart domain image from the target lung domain image according to the heart position information.
Optionally, the enhancement processing module is further configured to:
and taking an image formed by pixel points of which the parameter values of the preset parameters in the tubular domain enhanced image are greater than or equal to the preset blood vessel parameter threshold value as the blood vessel domain enhanced image.
Optionally, the enhancement processing module is further configured to:
determining a plurality of vessel domain connected images corresponding to the vessel domain enhanced image;
determining a plurality of target blood vessel domain connected images from the blood vessel domain connected images according to the preset position information;
and taking the union of a plurality of target blood vessel domain connected images as the coronary enhanced image.
Optionally, the segmentation module is further configured to:
determining a main coronary artery image corresponding to the coronary artery prediction image;
determining a plurality of coronary vessel domain connected images corresponding to the coronary artery enhanced image;
and according to the coronary artery main branch image and the coronary artery vessel domain communication images, carrying out segmentation processing on the original image to obtain the coronary artery segmentation image.
Optionally, the segmentation module is further configured to:
taking the largest coronary vessel domain connected image in the plurality of coronary vessel domain connected images as a target connected image, and circularly executing an image processing step according to the target connected image and the coronary artery main branch image until each coronary vessel domain connected image executes the image processing step, and taking a new coronary artery main branch image as the coronary artery segmentation image;
wherein the image processing step comprises:
determining an intersection image of the target connected image and the main coronary artery image;
determining whether a connected domain exists in the intersection image;
taking the union of the target connected image and the main coronary artery image as a new main coronary artery image and the largest connected coronary artery image in the rest connected coronary artery images as a new target connected image under the condition that the intersection images have connected domains, wherein the rest connected coronary artery images comprise the connected coronary artery images which are not subjected to the image processing step in the plurality of connected coronary artery images;
and under the condition that the intersection images do not have the connected image, taking the largest coronary vessel region connected image in the residual coronary vessel region connected images as a new target connected image.
In a third aspect, the present disclosure provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the method of the first aspect of the present disclosure.
In a fourth aspect, the present disclosure provides an electronic device comprising: a memory having a computer program stored thereon; a processor for executing the computer program in the memory to implement the steps of the method of the first aspect of the disclosure.
According to the technical scheme, the original image containing the heart blood vessel region is obtained; inputting the original image into a pre-trained coronary artery prediction model to obtain a coronary artery prediction image output by the coronary artery prediction model, wherein the coronary artery prediction image is used for predicting coronary arteries in the cardiovascular region; carrying out image enhancement processing on the heart blood vessel region in the original image to obtain a coronary artery enhanced image; and according to the coronary artery prediction image and the coronary artery enhanced image, carrying out segmentation processing on the original image to obtain a coronary artery segmentation image corresponding to the original image. That is, according to the present disclosure, a coronary artery in an original image is first segmented by a coronary artery prediction model to obtain a predicted coronary artery image, then an image enhancement process is performed on a cardiac vessel region in the original image to obtain a enhanced coronary artery image, and finally, a coronary artery in the original image is segmented by combining the predicted coronary artery image and the enhanced coronary artery image, so that the coronary artery in the predicted coronary artery image is repaired by the enhanced coronary artery image, and a more accurate coronary artery segmentation result can be obtained, thereby improving the effect of coronary artery segmentation.
Additional features and advantages of the disclosure will be set forth in the detailed description which follows.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure and are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure without limiting the disclosure. In the drawings:
FIG. 1 is a flow chart illustrating a method of coronary artery segmentation according to an exemplary embodiment of the present disclosure;
FIG. 2 is a flow chart illustrating another method of coronary artery segmentation according to an exemplary embodiment of the present disclosure;
FIG. 3 is a flow chart illustrating another method of coronary artery segmentation according to an exemplary embodiment of the present disclosure;
FIG. 4 is a flow chart illustrating another method of coronary artery segmentation according to an exemplary embodiment of the present disclosure;
FIG. 5 is a flow chart illustrating another method of coronary artery segmentation according to an exemplary embodiment of the present disclosure;
FIG. 6 is a flowchart illustrating one image processing step according to an exemplary embodiment of the present disclosure;
FIG. 7 is a contrast diagram of a coronary artery shown in accordance with an exemplary embodiment of the present disclosure;
FIG. 8 is a block diagram illustrating a coronary artery segmentation apparatus according to an exemplary embodiment of the present disclosure;
FIG. 9 is a block diagram illustrating another coronary artery segmentation apparatus in accordance with an exemplary embodiment of the present disclosure;
fig. 10 is a block diagram illustrating an electronic device in accordance with an exemplary embodiment of the present disclosure.
Detailed Description
The following detailed description of specific embodiments of the present disclosure is provided in connection with the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating the present disclosure, are given by way of illustration and explanation only, not limitation.
It should be noted that all the actions of acquiring signals, information or data in the present disclosure are performed under the premise of complying with the corresponding data protection regulation policy of the country of the location and obtaining the authorization given by the owner of the corresponding device.
In the description that follows, the terms "first," "second," and the like are used for descriptive purposes only and are not intended to indicate or imply relative importance nor order to be construed.
First, an application scenario of the present disclosure will be explained. In the early stage of coronary artery segmentation, the coronary artery is mostly extracted based on the coronary artery vessel characteristics by adopting an image processing technology, however, the method needs to manually set parameters such as various position information and thresholds, the automation degree is low, and the accuracy of coronary artery segmentation is influenced. In the related art, the coronary artery is segmented by a deep learning technique, although the segmentation result of the coronary artery can be directly obtained. However, since the coronary artery vessels are relatively thin, particularly, the terminal capillaries, it is difficult to label the terminal capillaries in the sample data, and the coronary artery segmentation model cannot be learned without labeled data, the coronary artery segmentation method based on the deep learning technique can only segment the coronary artery main vessels and the branch vessels accurately, and the effect of segmenting the terminal capillaries is relatively poor. In addition, the segmentation result of the coronary artery based on the deep learning technique may have fragments such as scattered vein vessels, which affect the segmentation effect of the coronary artery, and may cause a fracture phenomenon in the segmentation result of the coronary artery if the coronary artery vessel is diseased.
In order to solve the above problems, the present disclosure provides a coronary artery segmentation method, apparatus, storage medium, and electronic device, wherein a coronary artery in an original image is segmented by a coronary artery prediction model to obtain a predicted coronary artery image, then an image enhancement process is performed on a cardiac blood vessel region in the original image to obtain a enhanced coronary artery image, and finally a coronary artery in the original image is segmented by combining the predicted coronary artery image and the enhanced coronary artery image, so that the coronary artery in the predicted coronary artery image is repaired by the enhanced coronary artery image, and a more accurate coronary artery segmentation result can be obtained, thereby improving the effect of coronary artery segmentation.
Specific embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.
Fig. 1 is a flowchart illustrating a coronary artery segmentation method according to an exemplary embodiment of the present disclosure, which may include, as shown in fig. 1:
s101, acquiring an original image containing a heart blood vessel area.
The raw image may be a cardiac CTA (Computed Tomography Angiography) image, among others. Since the heart is located between the two lungs, the CTA image will typically include a heart region, a portion of, or all of the lung region.
In this step, the original image may be obtained by performing a Tomography scan on a heart region of the human body through a CT (Computed Tomography) apparatus, and the manner of obtaining the original image is only an example, and the disclosure is not limited thereto.
And S102, inputting the original image into a pre-trained coronary artery prediction model to obtain a coronary artery prediction image output by the coronary artery prediction model.
Wherein the coronary predicted image can be used to predict coronary arteries in the cardiovascular region.
In this step, after the original image is obtained, the original image may be input into the coronary artery prediction model, and the coronary artery prediction image is obtained by labeling the blood vessel region and the non-blood vessel region in the original image with the coronary artery prediction model. The coronary artery prediction model may be a model trained in the prior art for segmenting coronary arteries, or may be obtained by training a model trained in the prior art, which is not described herein again.
S103, carrying out image enhancement processing on the heart blood vessel region in the original image to obtain a coronary enhanced image.
In this step, a heart region may be extracted from the original image, and an image enhancement process is performed on a cardiovascular region in the heart region by using an image enhancement technique in the prior art, so as to obtain the coronary enhanced image.
And S104, according to the coronary predicted image and the coronary enhanced image, carrying out segmentation processing on the original image to obtain a coronary segmentation image corresponding to the original image.
In this step, after obtaining the predicted coronary artery image and the enhanced coronary artery image, a main coronary artery image corresponding to the predicted coronary artery image may be determined, a plurality of coronary artery domain connected images corresponding to the enhanced coronary artery image may be determined, and the main coronary artery image and the plurality of coronary artery domain connected images may be fused to repair the marked coronary artery in the coronary artery image, so as to obtain a segmented coronary artery image corresponding to the original image.
In summary, according to the present disclosure, a coronary artery in an original image is first segmented by a coronary artery prediction model to obtain a predicted coronary artery image, then an image enhancement process is performed on a cardiac vessel region in the original image to obtain a enhanced coronary artery image, and finally, a coronary artery in the original image is segmented by combining the predicted coronary artery image and the enhanced coronary artery image, so that the coronary artery in the predicted coronary artery image is repaired by the enhanced coronary artery image, and a more accurate coronary artery segmentation result can be obtained, thereby improving the effect of coronary artery segmentation.
Fig. 2 is a flowchart illustrating another coronary artery segmentation method according to an exemplary embodiment of the present disclosure, and as shown in fig. 2, the implementation of step S103 may include:
and S1031, determining a blood vessel region enhanced image containing the heart blood vessel region according to the original image.
For example, a heart domain image containing only a heart region may be determined from the original image, and then the blood vessel region enhancement image may be obtained by performing image enhancement processing on the blood vessel region of the heart in the heart domain image by using an image enhancement processing method in the prior art.
S1032, determining the coronary artery enhanced image according to the preset position information and the blood vessel domain enhanced image.
Wherein the preset position information may be position information of a preset coronary artery region in the original image. The present disclosure may analyze a plurality of historical CTA images, determine, for each historical CTA image, position information of a coronary artery in the historical CTA image, and finally determine the preset position information according to the plurality of position information. It should be noted that the preset position information may be different from the original image acquired by different devices, and the disclosure does not limit this.
Fig. 3 is a flowchart illustrating another coronary artery segmentation method according to an exemplary embodiment of the present disclosure, and as shown in fig. 3, the implementation of step S1031 may include:
s31, a heart region image including the heart region is determined based on the original image.
The original image includes both the heart region and the lung region, and the CT value of the lung region in the original image is relatively low, generally between-900 Hu to-350 Hu, due to the characteristics of the lungs. For the above reasons, before determining the heart domain image, a lung domain image including a lung region may be determined from the original image, and the heart domain image may be extracted from the lung domain image according to preset heart position information. The heart position information may be preset empirically, and for example, the heart position information may include first position information of the heart region in a first direction, which may be an x-axis direction, and second position information of the heart in a second direction, which may be a y-axis direction, the first position information is used to represent a distance of the heart region from two sides of the lung field image in the first direction, for example, the first position information may be 50 pixel points, and the second position information is used to represent a distance of the heart region from two sides of the lung field image in the second direction, for example, the second position information may be 20 pixel points.
The method includes determining a lung domain binary image corresponding to the original image according to a preset lung domain parameter threshold, determining a plurality of lung connected domain images in the lung domain binary image, and determining the lung domain image according to the plurality of lung connected domain images. The preset lung region parameter threshold may be a CT value range corresponding to a lung region, for example, the preset lung region parameter threshold may be-900 Hu to-350 Hu.
Exemplarily, for each pixel point in the original image, a CT value corresponding to the pixel point is determined, and an image formed by target pixel points whose CT values are within the preset lung domain parameter threshold range is used as the lung domain binary image. Then, the lung domain binary image may be divided according to the position information of each pixel point in the lung domain binary image to obtain a plurality of lung connected domain images, for example, for each pixel point in the lung domain binary image, whether there is an adjacent pixel point for the pixel point may be determined, if there is no adjacent pixel point for the pixel point, the pixel point is used as a boundary point of a lung connected domain, and finally, a plurality of lung connected domain images are obtained after connecting the plurality of boundary points.
Further, after obtaining a plurality of lung connected domain images, a first length value corresponding to a first lung connected domain image may be determined, where the first lung connected domain image is a largest lung connected domain image in the plurality of lung connected domain images, and the first length value is a length of the first lung connected domain image in a first direction, for example, the first direction may be an x-axis direction of the original image, and the length may be the number of pixel points; when the difference between the first length value and the second length value is greater than or equal to a preset difference threshold, the first lung connected domain image is taken as the lung domain image, the second length value is the length of the original image in the first direction, and the preset difference threshold may be a determination margin, for example, the preset difference threshold may be 20. And under the condition that the difference value between the first length value and the second length value is smaller than the preset difference threshold value, splicing the first lung connected domain image and a second lung connected domain image to obtain the lung domain image, wherein the second lung connected domain image is the largest lung connected domain image except the first lung connected domain image in the plurality of lung connected domain images.
For example, the size of each lung connected domain image may be determined first, where the size may be the number of pixel points included in the lung connected domain image, and then the lung connected domain images are sorted in descending order, and the largest lung connected domain image in the lung connected domain images is taken as the first lung connected domain image. Then, the length of the first lung connected domain image in the x axis may be determined to obtain the first length value, the length of the original image in the x axis may be determined to obtain the second length value, a difference between the first length value and the second length value is determined, and if the difference is greater than or equal to the preset difference threshold, the first lung connected domain image is determined to be the lung domain image.
If the difference is smaller than or equal to the preset difference threshold, the first lung connected domain image is represented as a half lung domain, then, the second lung connected domain image is determined from the plurality of lung connected domain images, the second lung connected domain image is the other half lung domain, the first lung connected domain image and the second lung connected domain image are subjected to union operation, and the lung domain image is obtained through splicing.
After obtaining the lung region image, obtaining preset heart position information, and performing retraction processing on the lung region image according to the heart position information, for example, if the first position information in the heart position information is 50 pixel points, and the second position information is 20 pixel points, both sides of the lung region image along the x-axis direction can be retracted inward by 50 pixel points, both sides of the lung region image along the y-axis direction are retracted inward by 20 pixel points, so as to obtain an undetermined heart region image, and then performing complementary operation on the lung region image and the undetermined heart region image, so as to remove redundant lung regions in the undetermined heart region image, so as to obtain a heart region image only including heart regions.
It should be noted that, before the heart domain image is extracted from the lung domain image, the holes in the lung domain image may be filled to obtain a target lung domain image, and then the heart domain image is extracted from the target lung domain image according to the heart position information. For example, the holes in the lung domain image may be filled by a morphological closing operation, and the kernel radius for determining the circular convolution kernel used in the morphological closing operation may take a larger value, for example, the kernel radius may be 8, so as to ensure that the holes in the lung domain image can be filled.
And S32, performing enhancement processing on the heart domain image to obtain a heart domain enhanced image.
After the heart domain image is determined, the boundary of the tubular region in the heart domain image can be enhanced through a Hessian _ recursion _ gaussian image filtering function in the prior art to obtain the heart domain Hessian image, wherein the value of a sigma parameter corresponding to the image filtering function can be 0.8. Then, the heart domain Hessian image can be processed through a Hessian3dtovesselness measure image filtering function in the prior art, so as to measure the tubular object in the heart domain Hessian image, and obtain a tubular domain enhanced image. Finally, 2-class segmentation (background and foreground) is carried out on the enhanced image of the tubular domain through an Otsu image automatic segmentation algorithm, a non-tubular region (background) in the enhanced image of the tubular domain is removed, and a heart domain enhanced image only containing the tubular domain is obtained, wherein the heart domain enhanced image is a 2-value image.
S33, the heart region enhanced image is segmented to obtain a tube region enhanced image including a tube region.
After obtaining the heart domain enhanced image, the heart domain enhanced image may be multiplied by the original image to obtain a tube domain enhanced image containing only tube regions.
And S34, determining the enhanced image of the blood vessel domain according to the preset blood vessel parameter threshold value and the enhanced image of the tubular domain.
Wherein the preset blood vessel parameter threshold may be a CT value of a blood vessel region, and exemplarily, the preset blood vessel parameter threshold may be-200 Hu.
After the enhanced image of the tubular region is obtained, the enhanced image of the tubular region can be segmented by a threshold segmentation method in the prior art to obtain the enhanced image of the blood vessel region. In a possible implementation manner, an image composed of pixels whose parameter values of the preset parameters in the tubular domain enhanced image are greater than or equal to the preset blood vessel parameter threshold may be used as the blood vessel domain enhanced image. The preset parameter may be a CT value, for example, the preset blood vessel parameter threshold may be obtained, and an image composed of pixel points in the tubular domain enhanced image whose CT value is greater than or equal to the preset blood vessel parameter threshold is used as the blood vessel domain enhanced image.
Fig. 4 is a flowchart illustrating another coronary artery segmentation method according to an exemplary embodiment of the present disclosure, and as shown in fig. 4, the implementation of step S1032 may include:
and S41, determining a plurality of blood vessel domain connected images corresponding to the blood vessel domain enhanced image.
After obtaining the blood vessel region enhanced image, referring to the manner of determining the multiple lung connected domain images in step S31, multiple blood vessel region connected images corresponding to the blood vessel region enhanced image may be determined, which is not described herein again.
S42, determining a plurality of target blood vessel region connected images from the plurality of blood vessel region connected images according to the preset position information.
For example, the preset position information may include abscissa information in an x-axis direction and ordinate information in a y-axis direction, the abscissa information may include an initial abscissa range and an end abscissa threshold, the ordinate information may also include an initial ordinate threshold and an end ordinate threshold, if the resolution of the original image is 512 × 512, the initial abscissa range may be [100, 350], the end abscissa threshold may be 412, the initial ordinate threshold may be 300, and the end ordinate threshold may be 450.
After obtaining the plurality of blood vessel domain connected images, the preset position information may be obtained, and a target blood vessel domain connected image of which the position information satisfies the preset position information is determined from the plurality of blood vessel domain connected images. For example, an initial abscissa, an end abscissa, an initial ordinate and an end ordinate of each blood vessel domain connected image may be determined, then a plurality of pending blood vessel domain connected images with an initial abscissa within 100-350 and an end abscissa smaller than 412 are determined from the plurality of blood vessel domain connected images, and finally, a target pending blood vessel domain connected image with an initial ordinate smaller than 300 and an end ordinate smaller than 450 is determined from the plurality of pending blood vessel domain connected images, and the target pending blood vessel domain connected image is taken as the target blood vessel domain connected image.
S43, using the union of the target blood vessel region connected images as the coronary artery enhanced image.
After obtaining the multiple target vessel domain connected images, union operation may be performed on the multiple target vessel domain connected images to obtain the coronary artery enhanced image.
Fig. 5 is a flowchart illustrating another coronary artery segmentation method according to an exemplary embodiment of the present disclosure, and as shown in fig. 5, the implementation of step S104 may include:
and S1041, determining a main coronary artery image corresponding to the coronary artery prediction image.
After obtaining the predicted coronary artery image, the plurality of predicted connected images corresponding to the predicted coronary artery image may be determined in a manner of determining the plurality of pulmonary connected region images in step S31, and the description thereof is omitted here.
Further, the multiple prediction connected images may be arranged in an order from large to small, two maximum prediction connected images, namely a first prediction connected image and a second prediction connected image, in the multiple prediction connected images are determined, and then the first prediction connected image and the second prediction connected image are subjected to union operation to obtain a main coronary artery image containing the two main coronary arteries, wherein the main coronary artery image is a binary image.
And S1042, determining a plurality of coronary artery blood vessel domain connected images corresponding to the coronary artery enhanced image.
After obtaining the coronary artery enhanced image, the coronary artery vessel region connected images corresponding to the coronary artery enhanced image may be determined by referring to the manner of determining the lung connected region images in step S31, which is not described herein again.
And S1043, according to the coronary artery main branch image and the multiple coronary artery vessel domain communication images, performing segmentation processing on the original image to obtain the coronary artery segmentation image.
In a possible implementation manner, the largest coronary vessel region connected image in a plurality of coronary vessel region connected images can be used as a target connected image, and according to the target connected image and the coronary artery main branch image, the image processing step is executed in a loop until each coronary vessel region connected image executes the image processing step, and a new coronary artery main branch image is used as the coronary artery segmentation image.
Fig. 6 is a flowchart illustrating an image processing step according to an exemplary embodiment of the present disclosure, which may include, as shown in fig. 6:
and S61, determining an intersection image of the target connected image and the main coronary artery image.
S62, determining whether there is a connected domain in the intersection image, if there is a connected domain in the intersection image, executing step S63, and if there is no connected domain in the intersection image, executing step S64.
S63, regarding the union of the target connected image and the main coronary artery image as a new main coronary artery image, and regarding the largest coronary artery connected image in the remaining coronary artery connected images as a new target connected image, where the remaining coronary artery connected images include a plurality of coronary artery connected images in the coronary artery connected images for which the image processing step is not performed.
S64, the largest coronary vessel region connected image among the remaining coronary vessel region connected images is set as a new target connected image.
For example, after determining the main coronary artery map and the multiple coronary artery domain connected images, the multiple coronary artery domain connected images may be sorted in descending order to obtain an image sequence, a first coronary artery domain connected image in the image sequence is taken as the target connected image, and the target connected image and the main coronary artery map are subjected to intersection operation to obtain an intersection image of the target connected image and the main coronary artery map. Then, determining whether a connected domain exists in the intersection image, if the connected domain exists in the intersection image, performing union operation on the target connected image and the coronary artery main branch image to obtain a new coronary artery main branch image, selecting a second coronary artery blood vessel domain connected image from the image sequence, and taking the second coronary artery blood vessel domain connected image as a new target connected image; if the intersection image does not have a connected region, selecting a second coronary vessel region connected image from the image sequence, and taking the second coronary vessel region connected image as a new target connected image; if the intersection image has no connected component, the main coronary artery image is not updated.
And circularly executing the image processing steps for each coronary vessel domain connected image in the image sequence until each coronary vessel domain connected image in the image sequence is executed, and taking the finally determined new main coronary artery image as the coronary artery segmentation image.
Fig. 7 is a coronary artery contrast diagram according to an exemplary embodiment of the disclosure, as shown in fig. 7, the left side is a coronary artery prediction image, and the right side is the coronary artery segmentation image, in the coronary artery prediction image, there is a missing coronary artery end branch thin blood vessel and there are some scattered small vein blood vessel fragments, the coronary artery end branch thin blood vessel in the coronary artery segmentation image obtained after the repair by the scheme of the disclosure has been repaired, as the position marked by the white circle in the figure, and the small vein blood vessel fragments are also removed.
In summary, for the predicted coronary artery image obtained by the coronary artery prediction model, the image enhancement processing is performed on the heart blood vessel region in the original image to obtain the enhanced coronary artery image, so as to obtain the segmented coronary artery image including the complete coronary artery blood vessel tree.
Fig. 8 is a block diagram illustrating a coronary artery segmentation apparatus according to an exemplary embodiment of the present disclosure, which, as shown in fig. 8, may include:
a first acquisition module 801 for acquiring an original image containing a cardiovascular region;
a second obtaining module 802, configured to input the original image into a pre-trained coronary artery prediction model to obtain a coronary artery prediction image output by the coronary artery prediction model, where the coronary artery prediction image is used to predict coronary arteries in the cardiovascular region;
an enhancement processing module 803, configured to perform image enhancement processing on the cardiovascular region in the original image to obtain a coronary enhanced image;
a segmentation module 804, configured to perform segmentation processing on the original image according to the predicted coronary artery image and the enhanced coronary artery image, so as to obtain a segmented coronary artery image corresponding to the original image.
Optionally, the enhancement processing module 803 is further configured to:
determining a blood vessel domain enhanced image containing the heart blood vessel region according to the original image;
and determining the coronary artery enhanced image according to preset position information and the blood vessel domain enhanced image, wherein the preset position information is preset position information of a coronary artery region in the original image.
Optionally, the enhancement processing module 803 is further configured to:
determining a heart domain image containing a heart region from the original image;
performing enhancement processing on the heart domain image to obtain a heart domain enhanced image;
carrying out segmentation processing on the heart domain enhanced image to obtain a tubular domain enhanced image containing a tubular region;
and determining the vessel domain enhanced image according to a preset vessel parameter threshold and the tubular domain enhanced image.
Optionally, the enhancement processing module 803 is further configured to:
determining a lung area image containing a lung area according to the original image;
and extracting the heart domain image from the lung domain image according to preset heart position information.
Optionally, the enhancement processing module 803 is further configured to:
determining a lung area binary image corresponding to the original image according to a preset lung area parameter threshold;
determining a plurality of lung connected domain images in the lung domain binary image;
and determining the lung domain image according to a plurality of lung connected domain images.
Optionally, the enhancement processing module 803 is further configured to:
determining a first length value corresponding to a first lung connected domain image, wherein the first lung connected domain image is the largest lung connected domain image in a plurality of lung connected domain images, and the first length value is the length of the first lung connected domain image in a first direction;
and taking the first lung connected domain image as the lung domain image under the condition that the difference value between the first length value and a second length value is greater than or equal to a preset difference threshold value, wherein the second length value is the length of the original image in the first direction.
Optionally, the enhancement processing module 803 is further configured to:
and under the condition that the difference value between the first length value and the second length value is smaller than the preset difference threshold value, splicing the first lung connected domain image and a second lung connected domain image to obtain the lung domain image, wherein the second lung connected domain image is the largest lung connected domain image except the first lung connected domain image in the plurality of lung connected domain images.
Alternatively, fig. 9 is a block diagram illustrating another coronary artery segmentation apparatus according to an exemplary embodiment of the present disclosure, as shown in fig. 9, the apparatus further includes:
a filling module 805, configured to fill a hole in the lung region image to obtain a target lung region image;
the enhancement processing module 803 is further configured to:
and extracting the heart domain image from the target lung domain image according to the heart position information.
Optionally, the enhancement processing module is further configured to:
and taking an image formed by pixel points of which the parameter values of the preset parameters in the tubular domain enhanced image are greater than or equal to the preset blood vessel parameter threshold value as the blood vessel domain enhanced image.
Optionally, the enhancement processing module 803 is further configured to:
determining a plurality of blood vessel domain connected images corresponding to the blood vessel domain enhanced image;
determining a plurality of target blood vessel domain connected images from a plurality of blood vessel domain connected images according to the preset position information;
and taking a union of a plurality of target blood vessel domain connected images as the coronary enhanced image.
Optionally, the segmenting module 804 is further configured to:
determining a main coronary artery image corresponding to the predicted coronary artery image;
determining a plurality of coronary vessel domain connected images corresponding to the coronary artery enhanced image;
and according to the coronary artery main branch image and the plurality of coronary artery vessel domain communication images, carrying out segmentation processing on the original image to obtain the coronary artery segmentation image.
Optionally, the segmenting module 804 is further configured to:
taking the largest coronary vessel domain connected image in the plurality of coronary vessel domain connected images as a target connected image, and circularly executing an image processing step according to the target connected image and the coronary artery main branch image until each coronary vessel domain connected image executes the image processing step, and taking a new coronary artery main branch image as the coronary artery segmentation image;
wherein the image processing step comprises:
determining an intersection image of the target connected image and the main coronary artery image;
determining whether a connected domain exists in the intersection image;
taking the union of the target connected image and the main coronary artery image as a new main coronary artery image and the largest connected coronary artery image in the residual connected coronary artery images as a new target connected image under the condition that the intersection image has a connected domain, wherein the residual connected coronary artery images comprise a plurality of connected coronary artery images which are not subjected to the image processing step;
and in the case that the intersection image does not have the connected image, taking the largest coronary vessel region connected image in the residual coronary vessel region connected images as a new target connected image.
According to the device, firstly, the coronary artery in the original image is segmented through the coronary artery prediction model to obtain the predicted coronary artery image, then, the image enhancement processing is carried out on the heart blood vessel region in the original image to obtain the enhanced coronary artery image, and finally, the coronary artery in the original image is segmented through combining the predicted coronary artery image and the enhanced coronary artery image, so that the coronary artery in the predicted coronary artery image is repaired through the enhanced coronary artery image, a more accurate coronary artery segmentation result can be obtained, and the effect of coronary artery segmentation is improved.
With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.
Fig. 10 is a block diagram illustrating an electronic device 1000 in accordance with an exemplary embodiment of the present disclosure. For example, the electronic device 1000 may be provided as a server. Referring to fig. 10, the electronic device 1000 includes a processor 1022, which may be one or more in number, and a memory 1032 for storing computer programs executable by the processor 1022. The computer programs stored in memory 1032 may include one or more modules that each correspond to a set of instructions. Further, the processor 1022 may be configured to execute the computer program to perform the above-described coronary artery segmentation method.
Additionally, the electronic device 1000 may also include a power component 1026 and a communication component 1050, the power component 1026 may be configured to perform power management for the electronic device 1000, and the communication component 1050 may be configured to enable communication for the electronic device 1000, e.g., wired or wireless communication. In addition, the electronic device 1000 may also include input/output (I/O) interfaces 1058. The electronic device 1000 may operate based on an operating system stored in the memory 1032, such as Windows Server TM ，Mac OS X TM ，Unix TM ，Linux TM And so on.
In another exemplary embodiment, a computer readable storage medium comprising program instructions which, when executed by a processor, implement the steps of the above-described coronary artery segmentation method is also provided. For example, the non-transitory computer readable storage medium may be the memory 1032 comprising program instructions executable by the processor 1022 of the electronic device 1000 to perform the coronary artery segmentation method described above.
In another exemplary embodiment, a computer program product is also provided, which comprises a computer program executable by a programmable apparatus, the computer program having code portions for performing the above-mentioned coronary artery segmentation method when executed by the programmable apparatus.
The preferred embodiments of the present disclosure are described in detail with reference to the accompanying drawings, however, the present disclosure is not limited to the specific details of the above embodiments, and various simple modifications may be made to the technical solution of the present disclosure within the technical idea of the present disclosure, and these simple modifications all belong to the protection scope of the present disclosure. It should be noted that, in the foregoing embodiments, various features described in the above embodiments may be combined in any suitable manner, and in order to avoid unnecessary repetition, various combinations that are possible in the present disclosure are not described again.
In addition, any combination of various embodiments of the present disclosure may be made, and the same should be considered as the disclosure of the present disclosure, as long as it does not depart from the spirit of the present disclosure.

Claims (15)

1. A coronary artery segmentation method, comprising:
acquiring an original image containing a cardiovascular region;
inputting the original image into a pre-trained coronary artery prediction model to obtain a coronary artery prediction image output by the coronary artery prediction model, wherein the coronary artery prediction image is used for predicting coronary arteries in the cardiovascular region;
carrying out image enhancement processing on the heart blood vessel region in the original image to obtain a coronary artery enhanced image;
and according to the coronary artery prediction image and the coronary artery enhanced image, carrying out segmentation processing on the original image to obtain a coronary artery segmentation image corresponding to the original image.
2. The method of claim 1, wherein the image enhancement processing of the cardiovascular region in the original image to obtain a coronary enhanced image comprises:
determining a blood vessel domain enhanced image containing the heart blood vessel region according to the original image;
and determining the coronary artery enhanced image according to preset position information and the blood vessel domain enhanced image, wherein the preset position information is the position information of a preset coronary artery region in the original image.
3. The method of claim 2, wherein determining, from the original image, a vessel-domain-enhanced image containing the cardiovascular region comprises:
determining a heart domain image containing a heart region from the original image;
enhancing the heart domain image to obtain a heart domain enhanced image;
carrying out segmentation processing on the heart domain enhanced image to obtain a tubular domain enhanced image containing a tubular region;
and determining the blood vessel domain enhanced image according to a preset blood vessel parameter threshold value and the tubular domain enhanced image.
4. The method of claim 3, wherein determining a heart domain image containing a heart region from the raw image comprises:
determining a lung area image containing a lung area according to the original image;
and extracting the heart domain image from the lung domain image according to preset heart position information.
5. The method of claim 4, wherein determining a lung field image containing a lung region from the original image comprises:
determining a lung area binary image corresponding to the original image according to a preset lung area parameter threshold;
determining a plurality of lung connected domain images in the lung domain binary image;
and determining the lung domain image according to the lung connected domain images.
6. The method of claim 5, wherein determining the lung domain image from the plurality of lung connected domain images comprises:
determining a first length value corresponding to a first lung connected domain image, wherein the first lung connected domain image is the largest lung connected domain image in the plurality of lung connected domain images, and the first length value is the length of the first lung connected domain image in a first direction;
and taking the first lung connected domain image as the lung domain image when the difference between the first length value and a second length value is greater than or equal to a preset difference threshold, wherein the second length value is the length of the original image in the first direction.
7. The method of claim 6, further comprising:
and under the condition that the difference value between the first length value and the second length value is smaller than the preset difference threshold value, splicing the first lung connected domain image and the second lung connected domain image to obtain the lung domain image, wherein the second lung connected domain image is the largest lung connected domain image except the first lung connected domain image in the plurality of lung connected domain images.
8. The method according to claim 4, wherein before the extracting the heart domain image from the lung domain image according to the preset heart position information, the method further comprises:
filling holes in the lung area image to obtain a target lung area image;
the extracting the heart domain image from the lung domain image according to the preset heart position information comprises:
and extracting the heart domain image from the target lung domain image according to the heart position information.
9. The method according to claim 3, wherein said determining the vessel domain enhanced image according to the preset vessel parameter threshold and the tubular domain enhanced image comprises:
and taking an image formed by pixel points of which the parameter values of the preset parameters in the tubular domain enhanced image are greater than or equal to the preset blood vessel parameter threshold value as the blood vessel domain enhanced image.
10. The method according to claim 2, wherein the determining the coronary artery enhanced image according to the preset position information and the vessel domain enhanced image comprises:
determining a plurality of vessel domain connected images corresponding to the vessel domain enhanced image;
determining a plurality of target blood vessel domain connected images from the blood vessel domain connected images according to the preset position information;
and taking the union of a plurality of target blood vessel domain connected images as the coronary enhanced image.
11. The method according to any one of claims 1 to 10, wherein the performing segmentation processing on the original image according to the predicted coronary artery image and the enhanced coronary artery image to obtain a segmented coronary artery image corresponding to the original image comprises:
determining a main coronary artery image corresponding to the coronary artery prediction image;
determining a plurality of coronary vessel domain connected images corresponding to the coronary artery enhanced image;
and according to the coronary artery main branch image and the coronary artery vessel domain communication images, carrying out segmentation processing on the original image to obtain the coronary artery segmentation image.
12. The method according to claim 11, wherein the performing segmentation processing on the original image according to the main coronary artery map and the plurality of coronary vessel domain connected images to obtain the coronary artery segmentation image comprises:
taking the largest coronary vessel domain connected image in the plurality of coronary vessel domain connected images as a target connected image, circularly executing image processing steps according to the target connected image and the coronary artery main branch image until each coronary vessel domain connected image executes the image processing step, and taking a new coronary artery main branch image as the coronary artery segmentation image;
wherein the image processing step comprises:
determining an intersection image of the target connected image and the main coronary artery image;
determining whether a connected domain exists in the intersection image;
taking the union of the target connected image and the main coronary artery image as a new main coronary artery image and the largest connected coronary artery image in the rest connected coronary artery images as a new target connected image under the condition that the intersection images have connected domains, wherein the rest connected coronary artery images comprise the connected coronary artery images which are not subjected to the image processing step in the plurality of connected coronary artery images;
and under the condition that the intersection images do not have the connected image, taking the largest coronary vessel region connected image in the residual coronary vessel region connected images as a new target connected image.
13. A coronary artery segmentation device, comprising:
a first acquisition module for acquiring an original image containing a cardiovascular region;
a second obtaining module, configured to input the original image into a pre-trained coronary artery prediction model to obtain a coronary artery prediction image output by the coronary artery prediction model, where the coronary artery prediction image is used to predict coronary arteries in the cardiovascular region;
the enhancement processing module is used for carrying out image enhancement processing on the heart blood vessel region in the original image to obtain a coronary artery enhancement image;
and the segmentation module is used for carrying out segmentation processing on the original image according to the coronary predicted image and the coronary enhanced image to obtain a coronary segmentation image corresponding to the original image.
14. A non-transitory computer readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 12.
15. An electronic device, comprising:
a memory having a computer program stored thereon;
a processor for executing the computer program in the memory to carry out the steps of the method of any one of claims 1 to 12.
CN202210585956.7A 2022-05-26 2022-05-26 Coronary artery segmentation method, coronary artery segmentation device, storage medium, and electronic apparatus Pending CN114972257A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210585956.7A CN114972257A (en) 2022-05-26 2022-05-26 Coronary artery segmentation method, coronary artery segmentation device, storage medium, and electronic apparatus

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210585956.7A CN114972257A (en) 2022-05-26 2022-05-26 Coronary artery segmentation method, coronary artery segmentation device, storage medium, and electronic apparatus

Publications (1)

Publication Number Publication Date
CN114972257A true CN114972257A (en) 2022-08-30

Family

ID=82955844

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210585956.7A Pending CN114972257A (en) 2022-05-26 2022-05-26 Coronary artery segmentation method, coronary artery segmentation device, storage medium, and electronic apparatus

Country Status (1)

Country Link
CN (1) CN114972257A (en)

Similar Documents

Publication Publication Date Title
CN111145173B (en) Plaque identification method, device, equipment and medium of coronary angiography image
CN111145160B (en) Method, device, server and medium for determining coronary artery branches where calcified regions are located
CN112465834B (en) Blood vessel segmentation method and device
CN110110723B (en) Method and device for automatically extracting target area in image
CN111696089A (en) Arteriovenous determining method, device, equipment and storage medium
CN109741312A (en) A kind of Lung neoplasm discrimination method, device, equipment and medium
CN115409990B (en) Medical image segmentation method, device, equipment and storage medium
CN112037146A (en) Medical image artifact automatic correction method and device and computer equipment
CN117373070B (en) Method and device for labeling blood vessel segments, electronic equipment and storage medium
CN113469963B (en) Pulmonary artery image segmentation method and device
CN113889238B (en) Image identification method and device, electronic equipment and storage medium
CN113971728B (en) Image recognition method, training method, device, equipment and medium for model
CN115049807A (en) Method and device for establishing pulmonary blood vessel model and server
CN116245832B (en) Image processing method, device, equipment and storage medium
CN114972257A (en) Coronary artery segmentation method, coronary artery segmentation device, storage medium, and electronic apparatus
CN115482261A (en) Blood vessel registration method, device, electronic equipment and storage medium
CN112950734B (en) Coronary reconstruction method, device, electronic equipment and storage medium
CN115439453A (en) Vertebral body positioning method and device, electronic equipment and storage medium
CN114998582A (en) Coronary artery blood vessel segmentation method, device and storage medium
CN114187281A (en) Image processing method and device, electronic equipment and storage medium
CN115908413B (en) Contrast image segmentation method, electronic device, processing system and storage medium
CN115187598B (en) Method, apparatus, system, device and medium for processing angiography image
CN117274216B (en) Ultrasonic carotid plaque detection method and system based on level set segmentation
CN116128914A (en) Two-stage segmentation method for contrast image, electronic device, processing system and storage medium
CN115861255A (en) Model training method, device, equipment, medium and product for image processing

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