CN114782366B - Heart stent detection method and device, storage medium and electronic equipment - Google Patents

Heart stent detection method and device, storage medium and electronic equipment Download PDF

Info

Publication number
CN114782366B
CN114782366B CN202210429123.1A CN202210429123A CN114782366B CN 114782366 B CN114782366 B CN 114782366B CN 202210429123 A CN202210429123 A CN 202210429123A CN 114782366 B CN114782366 B CN 114782366B
Authority
CN
China
Prior art keywords
dimensional image
coronary artery
stent
labeling
region
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202210429123.1A
Other languages
Chinese (zh)
Other versions
CN114782366A (en
Inventor
赵冬冬
尹思源
王少康
陈宽
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.)
Infervision Medical Technology Co Ltd
Original Assignee
Infervision Medical Technology 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 Infervision Medical Technology Co Ltd filed Critical Infervision Medical Technology Co Ltd
Priority to CN202210429123.1A priority Critical patent/CN114782366B/en
Publication of CN114782366A publication Critical patent/CN114782366A/en
Application granted granted Critical
Publication of CN114782366B publication Critical patent/CN114782366B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • G06T19/00—Manipulating 3D models or images for computer graphics
    • G06T19/20—Editing of 3D images, e.g. changing shapes or colours, aligning objects or positioning parts
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/20—Image enhancement or restoration using local operators
    • G06T5/30—Erosion or dilatation, e.g. thinning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/73—Deblurring; Sharpening
    • 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
    • G06T7/60—Analysis of geometric attributes
    • G06T7/68—Analysis of geometric attributes of symmetry
    • 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
    • G06T2207/30101—Blood vessel; Artery; Vein; Vascular
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2219/00—Indexing scheme for manipulating 3D models or images for computer graphics
    • G06T2219/004—Annotating, labelling
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Landscapes

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

Abstract

The application relates to the technical field of image processing, in particular to a method and a device for detecting a cardiac stent, a storage medium and electronic equipment, and solves the problem that the efficiency of finding the cardiac stent from a chest area three-dimensional image through naked eye observation is very low. According to the heart stent detection method, a coronary artery labeling three-dimensional image to be detected is obtained, the coronary artery labeling three-dimensional image to be detected contains labeled coronary artery information, and then stent information in the coronary artery labeling three-dimensional image to be detected is determined based on the coronary artery labeling three-dimensional image to be detected and a stent detection model. Because the heart stent is positioned in the coronary blood vessel, the coronary artery labeling three-dimensional image to be detected, which is labeled with the coronary blood information, is input into the stent detection model, so that the stent detection model can refer to the coronary blood information to determine the position of the stent in the coronary blood vessel, and the efficiency and the accuracy of detecting the heart stent are improved.

Description

Heart stent detection method and device, storage medium and electronic equipment
Technical Field
The application relates to the technical field of image processing, in particular to a heart stent detection method, a heart stent detection device, a computer readable storage medium and electronic equipment.
Background
The heart stent is also called coronary artery stent, is a medical apparatus commonly used in heart intervention operation, and has the function of dredging arterial blood vessel. After the cardiac stent is placed in an arterial vessel, whether the cardiac stent is re-blocked or not at an irregular time is required to be checked. Thus, the physician needs to determine the heart stent position in the chest computerized tomography (Computed Tomography, CT) image to see if a reocclusion of the heart stent position occurs. Currently, doctors need to determine the cardiac stent position by visually observing chest CT images. However, the heart stent is small, and the heart stent is generally 2mm-5mm in diameter and 13mm-38mm in length. Therefore, finding the cardiac stent from the chest CT image by visual inspection is time consuming and labor consuming and inefficient.
Disclosure of Invention
In view of this, embodiments of the present application provide a cardiac stent detection method and cardiac stent detection apparatus, and a computer-readable storage medium and an electronic device, which solve the problem that the efficiency of finding a cardiac stent from a three-dimensional image of a chest region by visual observation is very low.
In a first aspect, an embodiment of the present application provides a method for detecting a cardiac stent, including: obtaining a coronary artery labeling three-dimensional image to be detected, wherein the coronary artery labeling three-dimensional image to be detected contains labeled coronary artery information; and determining stent information in the coronary artery labeling three-dimensional image to be detected based on the coronary artery labeling three-dimensional image to be detected and a stent detection model, wherein the stent detection model is used for detecting the coronary artery labeling three-dimensional image to be detected so as to determine the stent information in the coronary artery labeling three-dimensional image to be detected.
With reference to the first aspect of the present application, in some embodiments, obtaining a three-dimensional image of a coronary annotation to be detected includes: performing central line extraction operation on coronary artery information contained in the three-dimensional image of the heart region to obtain a first coronary artery central line; performing morphological expansion treatment on the first coronary artery central line to obtain a second coronary artery central line; and labeling the heart region three-dimensional image based on the second coronary artery central line, and obtaining the coronary artery labeling three-dimensional image to be detected.
With reference to the first aspect of the present application, in some embodiments, labeling a three-dimensional image of a heart region based on a second coronary centerline, obtaining a coronary labeling three-dimensional image to be detected includes: performing Gaussian blur processing on the second coronary artery central line to obtain a third coronary artery central line; and labeling the heart region three-dimensional image based on the third coronary artery central line, and obtaining the coronary artery labeling three-dimensional image to be detected.
In combination with the first aspect of the present application, in some embodiments, before performing a centerline extraction operation on coronary information included in the three-dimensional image of the heart region to obtain a first coronary centerline, the method further includes: and dividing the chest region three-dimensional image to obtain a heart region three-dimensional image, wherein the heart region three-dimensional image is a part of the chest region three-dimensional image.
In combination with the first aspect of the present application, in some embodiments, segmenting the three-dimensional image of the chest region to obtain a three-dimensional image of the heart region includes: the method comprises the steps of performing dicing processing on a three-dimensional chest image to obtain a plurality of three-dimensional chest image dices, wherein an overlapping area is formed between adjacent three-dimensional chest image dices in the plurality of three-dimensional chest image dices; a heart region three-dimensional image is determined based on the plurality of three-dimensional chest image patches and a heart region detection model, wherein the heart region detection model is used to detect heart regions in the plurality of three-dimensional chest image patches to determine the heart region three-dimensional image.
With reference to the first aspect of the present application, in some embodiments, determining stent information in a three-dimensional image to be detected based on the coronary labeling three-dimensional image to be detected and a stent detection model includes: determining a plurality of coronary artery labeling cut blocks to be detected based on the coronary artery labeling three-dimensional image to be detected; determining stent region information corresponding to each of the coronary labeling cut pieces to be detected based on the coronary labeling cut pieces to be detected and the stent detection model, wherein the stent information in the coronary labeling three-dimensional image to be detected comprises the stent region information corresponding to each of the coronary labeling cut pieces to be detected;
With reference to the first aspect of the present application, in some embodiments, after determining stent region information corresponding to each of the plurality of coronary labeling cuts to be detected based on the plurality of coronary labeling cuts to be detected and the stent detection model, the method further includes: mapping stent region information corresponding to each of a plurality of coronary artery labeling cut blocks to be detected to a heart region three-dimensional image so as to determine a stent region in the heart region three-dimensional image; or mapping the stent region information corresponding to each of the coronary artery labeling cut pieces to be detected to the chest region three-dimensional image so as to determine the stent region in the chest region three-dimensional image.
In combination with the first aspect of the application, in some embodiments, determining a three-dimensional image of the heart region based on the plurality of three-dimensional chest image patches and the heart region detection model comprises: determining a heart region in the three-dimensional chest region image based on the plurality of three-dimensional chest image patches and the heart region detection model; and intercepting a heart region in the chest region three-dimensional image by using an circumscribed cuboid algorithm so as to determine the heart region three-dimensional image.
With reference to the first aspect of the present application, in some embodiments, before determining stent information in the coronary labeling three-dimensional image to be detected based on the coronary labeling three-dimensional image to be detected and the stent detection model, the method further includes: constructing an initial network model and a loss function of the initial network model, wherein in the training process, the loss value of the loss function is determined based on coronary region information in the three-dimensional image sample; based on the three-dimensional image sample, coronary labeling information and stent labeling information corresponding to the three-dimensional image sample, and the loss function, training an initial network model to generate a stent detection model.
In a second aspect, an embodiment of the present application provides a cardiac stent detection device, including: the acquisition module is configured to acquire a coronary artery labeling three-dimensional image to be detected, wherein the coronary artery labeling three-dimensional image to be detected contains labeled coronary artery information; the detection module is configured to determine stent information in the coronary artery labeling three-dimensional image to be detected based on the coronary artery labeling three-dimensional image to be detected and a stent detection model, wherein the stent detection model is used for detecting the coronary artery labeling three-dimensional image to be detected so as to determine the stent information in the coronary artery labeling three-dimensional image to be detected.
In a third aspect, an embodiment of the present application provides a computer-readable storage medium storing instructions that, when executed by a processor of an electronic device, enable the electronic device to perform the cardiac stent detection method mentioned in the first aspect above.
In a fourth aspect, an embodiment of the present application provides an electronic device, including: a processor; a memory for storing computer-executable instructions; a processor for executing computer-executable instructions to implement the cardiac stent detection method mentioned in the first aspect above.
According to the heart stent detection method provided by the embodiment of the application, the coronary artery labeling three-dimensional image to be detected is obtained, the coronary artery labeling three-dimensional image to be detected contains labeled coronary artery information, and then the stent information in the coronary artery labeling three-dimensional image to be detected is determined based on the coronary artery labeling three-dimensional image to be detected and the stent detection model. Because the heart stent is positioned in the coronary blood vessel, coronary artery information contained in the three-dimensional image of the heart region is marked to obtain a coronary artery marked three-dimensional image to be detected, and then the coronary artery marked three-dimensional image to be detected, which is marked with the coronary artery information, is input into a stent detection model, so that the stent detection model can refer to the coronary artery information to determine the position of the stent in the coronary blood vessel, and the efficiency and the accuracy of detecting the heart stent are improved.
Drawings
Fig. 1 is a schematic diagram of an application scenario of a cardiac stent detection method according to an embodiment of the present application.
Fig. 2 is a flow chart of a method for detecting a cardiac stent according to an embodiment of the present application.
Fig. 3 is a schematic flow chart of a method for detecting a cardiac stent according to another embodiment of the present application.
Fig. 4 is a schematic flow chart of a method for detecting a cardiac stent according to another embodiment of the present application.
Fig. 5 is a schematic flow chart of a method for detecting a cardiac stent according to another embodiment of the present application.
Fig. 6 is a schematic flow chart of a method for detecting a cardiac stent according to another embodiment of the present application.
Fig. 7 is a schematic flow chart of a method for detecting a cardiac stent according to another embodiment of the present application.
Fig. 8 is a schematic flow chart of a method for detecting a cardiac stent according to another embodiment of the present application.
Fig. 9 is a schematic flow chart of a method for detecting a cardiac stent according to another embodiment of the present application.
Fig. 10 is a flowchart of a method for detecting a cardiac stent according to another embodiment of the present application.
Fig. 11 is a schematic structural diagram of a cardiac stent detection device according to an embodiment of the present application.
Fig. 12 is a schematic structural diagram of a cardiac stent detection device according to another embodiment of the present application.
Fig. 13 is a schematic structural diagram of a cardiac stent detection device according to another embodiment of the present application.
Fig. 14 is a schematic structural diagram of a cardiac stent detection device according to another embodiment of the present application.
Fig. 15 is a schematic structural diagram of a cardiac stent detection device according to another embodiment of the present application.
Fig. 16 is a schematic structural diagram of a cardiac stent detection device according to another embodiment of the present application.
Fig. 17 is a schematic structural diagram of a cardiac stent detection device according to another embodiment of the present application.
Fig. 18 is a schematic structural diagram of a cardiac stent detection device according to another embodiment of the present application.
Fig. 19 is a schematic structural diagram of a cardiac stent detection device according to another embodiment of the present application.
Fig. 20 is a schematic structural diagram of an electronic device according to an embodiment of the application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
Exemplary scenario
Fig. 1 is a schematic diagram of an application scenario of a cardiac stent detection method according to an embodiment of the present application. The scenario illustrated in fig. 1 includes a server 110 and an image generation device 120 communicatively coupled to the server 110. Specifically, the server 110 is configured to obtain a coronary labeling three-dimensional image to be detected, where the coronary labeling three-dimensional image to be detected includes labeled coronary information; and determining stent information in the coronary artery labeling three-dimensional image to be detected based on the coronary artery labeling three-dimensional image to be detected and a stent detection model, wherein the stent detection model is used for detecting the coronary artery labeling three-dimensional image to be detected so as to determine the stent information in the coronary artery labeling three-dimensional image to be detected. The image generating device 120 is configured to generate a three-dimensional image of the coronary artery annotation to be detected, and send the generated three-dimensional image of the coronary artery annotation to be detected to the server 110, so that the server 110 performs the above operation.
Exemplary method
Fig. 2 is a flow chart of a method for detecting a cardiac stent according to an embodiment of the present application. As shown in fig. 2, the method for detecting the cardiac stent comprises the following steps.
Step 210, obtaining a three-dimensional image of the coronary artery labeling to be detected.
Specifically, the coronary labeling three-dimensional image to be detected contains labeled coronary information. The obtaining of the coronary artery labeling three-dimensional image to be detected can be directly obtaining the coronary artery labeling three-dimensional image to be detected, which is already labeled with coronary artery information. The obtaining of the coronary artery labeling three-dimensional image to be detected can also be based on coronary artery information contained in the heart region three-dimensional image, and the coronary artery labeling three-dimensional image to be detected can be obtained by labeling the heart region three-dimensional image.
In particular, the three-dimensional image of the heart region may be a sequence of images of the heart site taken by CT. CT uses precisely collimated X-ray beam, gamma ray, ultrasonic wave, etc. to scan the cross section around a certain part of human body together with a detector with very high sensitivity. The coronary artery is an artery that supplies blood to the heart, and originates in the aortic sinus at the root of the aorta, and runs on the surface of the heart, almost around the heart for one week. The coronary information may be information of a coronary region contained in the three-dimensional image of the heart region. Based on coronary information contained in the three-dimensional image of the heart region, the three-dimensional image of the heart region is marked, and the coronary region can be marked in the three-dimensional image of the heart region to obtain a coronary marked three-dimensional image to be detected. The coronary artery labeling three-dimensional image to be detected can be obtained by labeling a three-dimensional coronary artery region in the heart region three-dimensional image.
The coronary region is marked in the three-dimensional image of the heart region, which can be automatic marking, for example, the coronary region in the three-dimensional image of the heart region is identified through a network model and marked, or the coronary region in the three-dimensional image of the heart region is marked by a binarization algorithm, a connected domain algorithm and the like, or the coronary region in the three-dimensional image of the heart region is marked manually. The labeling mode is not particularly limited as long as the coronary region can be labeled in the three-dimensional image of the heart region.
Step 220, determining stent information in the coronary artery labeling three-dimensional image to be detected based on the coronary artery labeling three-dimensional image to be detected and the stent detection model.
Specifically, the stent detection model is used for detecting the coronary artery labeling three-dimensional image to be detected so as to determine stent information in the coronary artery labeling three-dimensional image to be detected. The stent information in the coronary artery labeling three-dimensional image to be detected can be output by inputting the coronary artery labeling three-dimensional image to be detected into the stent detection model.
According to the heart stent detection method provided by the embodiment of the application, the coronary artery labeling three-dimensional image to be detected is obtained, the coronary artery labeling three-dimensional image to be detected contains labeled coronary artery information, and then the stent information in the coronary artery labeling three-dimensional image to be detected is determined based on the coronary artery labeling three-dimensional image to be detected and the stent detection model. Because the heart stent is positioned in the coronary blood vessel, coronary artery information contained in the three-dimensional image of the heart region is marked to obtain a coronary artery marked three-dimensional image to be detected, and then the coronary artery marked three-dimensional image to be detected, which is marked with the coronary artery information, is input into a stent detection model, so that the stent detection model can refer to the coronary artery information to determine the position of the stent in the coronary blood vessel, and the efficiency and the accuracy of detecting the heart stent are improved.
Fig. 3 is a schematic flow chart of a method for detecting a cardiac stent according to another embodiment of the present application. The embodiment shown in fig. 3 is extended from the embodiment shown in fig. 2, and differences between the embodiment shown in fig. 3 and the embodiment shown in fig. 2 are described in the following, and are not repeated.
As shown in fig. 3, in the embodiment of the present application, the step of obtaining a three-dimensional image of the coronary artery labeling to be detected includes the following steps.
Step 310, performing a center line extraction operation on coronary information included in the three-dimensional image of the heart region to obtain a first coronary center line.
Specifically, the centerline extraction operation may be to extract the centerline of the graph. For example, the center line of the rectangle is the center line along the longitudinal direction of the rectangle. The coronary information may be information of a coronary region contained in the three-dimensional image of the heart region. Therefore, the central line extraction operation for the coronary information included in the three-dimensional image of the heart region may be to extract the central line of the coronary vessel of the coronary region included in the three-dimensional image of the heart region, thereby obtaining the first coronary central line. That is, the first coronary centerline may be a centerline of a coronary vessel of a coronary region contained in the three-dimensional image of the heart region.
Step 320, performing morphological dilation processing on the first coronary artery centerline to obtain a second coronary artery centerline.
In particular, morphology refers to mathematical morphology in image processing. Mathematical morphology is an image analysis discipline based on lattice theory and topology, and is the basic theory of mathematical morphology image processing. The morphological basic operations include: binary corrosion and expansion, binary opening and closing operation, skeleton extraction, extreme corrosion, hit-miss transformation, morphological gradient, top-hat transformation, particle analysis, drainage basin transformation, gray value corrosion and expansion, gray value opening and closing operation, gray value morphological gradient and the like. The morphological dilation process may be a binary dilation in the morphological basic operation described above. The first coronary artery central line is subjected to morphological expansion treatment, so that the line width of the first coronary artery central line can be changed from small to large (i.e. from thin to thick). Thus, the line width of the second coronary centerline is greater than the line width of the first coronary centerline.
And 330, labeling the three-dimensional image of the heart region based on the second coronary artery central line, and obtaining a coronary artery labeling three-dimensional image to be detected.
Specifically, the marking of the heart region three-dimensional image based on the second coronary artery central line may be that the second coronary artery central line is marked in the heart region three-dimensional image, so as to obtain the coronary artery marked three-dimensional image to be detected. Namely, the coronary artery labeling three-dimensional image to be detected can be obtained by labeling the center line of the second coronary artery in the heart region three-dimensional image.
Since the cardiac stent is placed in the coronary vessel, the outer diameter of the cardiac stent is almost equal to the diameter of the coronary vessel, and thus erroneous judgment of the cardiac stent due to the similarity of the outer diameter of the cardiac stent and the diameter of the coronary vessel is easy. The coronary artery information is subjected to central line extraction operation to obtain a first coronary artery central line, then morphological expansion processing is carried out on the first coronary artery central line to obtain a second coronary artery central line, finally, a heart region three-dimensional image is marked on the basis of the second coronary artery central line to obtain a coronary artery marked three-dimensional image to be detected, the real width information of the coronary artery blood vessel can be hidden, the false judgment of the heart stent caused by the fact that the outer diameter of the heart stent is similar to the diameter of the coronary artery blood vessel is reduced, and the accuracy of the detected stent information is further improved.
Fig. 4 is a schematic flow chart of a method for detecting a cardiac stent according to another embodiment of the present application. The embodiment shown in fig. 4 is extended from the embodiment shown in fig. 3, and differences between the embodiment shown in fig. 4 and the embodiment shown in fig. 3 are described in detail, so that details of the differences will not be repeated.
As shown in fig. 4, in the embodiment of the present application, the step of obtaining the coronary artery labeling three-dimensional image to be detected based on the second coronary artery centerline labeling three-dimensional image of the heart region includes the following steps.
Step 410, performing a gaussian blur process on the second coronary artery centerline to obtain a third coronary artery centerline.
Specifically, the process of performing gaussian blur processing on an image is to convolve the image with a normal distribution. By performing the gaussian blur processing on the second coronary artery centerline, the second coronary artery centerline can be made more blurred. I.e. the blurring effect of the third coronary artery centerline is better than the blurring effect of the second coronary artery centerline.
And step 420, labeling the three-dimensional image of the heart region based on the third coronary artery central line, and obtaining a coronary artery labeling three-dimensional image to be detected.
Specifically, labeling the heart region three-dimensional image based on the third coronary artery central line, namely labeling the third coronary artery central line in the heart region three-dimensional image, so as to obtain the coronary artery labeling three-dimensional image to be detected. The coronary artery labeling three-dimensional image to be detected can be obtained by labeling the center line of the third coronary artery in the three-dimensional image of the heart region.
And (3) carrying out Gaussian blur processing on the second coronary artery central line to obtain a third coronary artery central line, and then labeling a heart region three-dimensional image based on the third coronary artery central line to obtain a coronary artery labeling three-dimensional image to be detected, so that coronary blood vessels can be blurred, the real width information of the coronary blood vessels can be further hidden, the false judgment of the cardiac stent caused by the fact that the outer diameter of the cardiac stent is similar to the diameter of the coronary blood vessels is reduced, and the accuracy of the detected stent information is further improved.
Fig. 5 is a schematic flow chart of a method for detecting a cardiac stent according to another embodiment of the present application. The embodiment shown in fig. 5 is extended from the embodiment shown in fig. 2, and differences between the embodiment shown in fig. 5 and the embodiment shown in fig. 2 are described in detail, so that details of the differences will not be repeated.
As shown in fig. 5, in the embodiment of the present application, before the step of determining stent information in the three-dimensional image of the heart region based on the three-dimensional image of the coronary artery labeling to be detected and the stent detection model, the following steps are further included.
Step 510, constructing an initial network model and a loss function of the initial network model, wherein a loss value of the loss function is determined based on the coronary region information in the three-dimensional image sample during the training process.
Specifically, when constructing the loss function of the initial network model, only the coronary region in the three-dimensional image sample is weighted, and no other region except the coronary region in the three-dimensional image sample is weighted. That is, the weights of the other regions in the three-dimensional image sample except the coronary region are all zero when constructing the loss function of the initial network model. During training of the initial network model, a loss value of the loss function is determined based on the coronary region information in the three-dimensional image sample. That is, when returning the loss value, the loss value is returned which is calculated only based on the coronary region information in the three-dimensional image sample and the loss function.
Step 520, training an initial network model based on the three-dimensional image sample, the coronary labeling information and the stent labeling information corresponding to the three-dimensional image sample, and the loss function to generate a stent detection model.
Specifically, training the initial network model may be supervised training of the initial network model. The initial network model may be a UNET model, or may be another network model, which is not specifically limited in the present application.
In the process of training the initial network model, the loss value of the loss function is determined based on the coronary region information in the three-dimensional image sample, so that the initial network model can concentrate on learning the coronary region information and the bracket position, and the convergence rate of the initial network model is improved.
In an embodiment of the present application, during the process of training the initial network model, resampling may be performed on the three-dimensional image sample in the Z-axis direction, and downsampling may be performed on the XY plane. The three-dimensional image sample may be a chest region image sequence consisting of a plurality of chest region image samples, the Z-axis being the direction perpendicular to the plurality of chest region image samples. The XY plane is a plane parallel to the plurality of chest region image samples. Illustratively, the resolution of the three-dimensional image sample may be 512×512 before the downsampling process, and 256×256 after the downsampling process. By resampling the three-dimensional image sample in the Z-axis direction and downsampling the three-dimensional image sample in the XY plane, the data volume of the three-dimensional image sample is reduced, and the training efficiency of the initial network model is improved. Illustratively, the three-dimensional image samples may be resampled to 48×256×256 at a resolution of the three-dimensional image samples.
In an embodiment of the present application, in the process of training the initial network model, a truncation process and a linear normalization process may be performed on HU values of the three-dimensional image sample. In particular, the HU value of the three-dimensional image sample may be truncated according to the window width level of the heart. The window width is the HU value range displayed on the three-dimensional image sample and the window level is the center position of the sliding window. The window width of the heart is typically 300HU-500HU and the window level is typically 30HU-50HU. The truncation process may be to set a background area, for example, a black area or a white area, according to HU values of the three-dimensional image sample. For example, a region smaller than 300HU may be set as a black region, and a region larger than 500HU may be set as a white region. After the HU value of the three-dimensional image sample is truncated, the HU value of the three-dimensional image sample is still larger, which is not beneficial to the convergence of the model, so that the HU value of the three-dimensional image sample is subjected to linear normalization processing, and the convergence speed of the heart region detection model is improved.
In an embodiment of the application, in the process of training an initial network model, the three-dimensional image sample can be diced to obtain a diced sample, and the diced sample is subjected to data enhancement so as to improve the generalization capability of training data. The data enhancement can be random rotation, random translation, random brightness change, local brightness change of the bracket and the like of the cut block. Other enhancement modes can be selected for data enhancement, and the application is not particularly limited.
Fig. 6 is a schematic flow chart of a method for detecting a cardiac stent according to another embodiment of the present application. The embodiment shown in fig. 6 is extended from the embodiment shown in fig. 2, and differences between the embodiment shown in fig. 6 and the embodiment shown in fig. 2 are described in detail, so that details of the differences will not be repeated.
As shown in fig. 6, in the embodiment of the present application, before the step of performing the center line extraction operation on the coronary artery information included in the three-dimensional image of the heart region to obtain the first coronary artery center line, the following steps are further included.
In step 610, the chest region three-dimensional image is segmented to obtain a heart region three-dimensional image.
Specifically, the heart region three-dimensional image is a part of the chest region three-dimensional image. The three-dimensional image of the chest region may be a sequence of images of the heart site taken by CT. The three-dimensional image of the heart region may be a three-dimensional image of the heart region segmented from the three-dimensional image of the chest region. The heart region three-dimensional image may be a region of interest (Region of Interest, ROI) in the chest region three-dimensional image.
In practical application, the image sequence taken by CT is a three-dimensional image of the chest region. Because the heart stent has a small duty ratio in the three-dimensional image of the chest region, the difficulty of directly detecting the heart stent in the three-dimensional image of the chest region is high. By dividing the chest region three-dimensional image, a heart region three-dimensional image is obtained, the occupation ratio of the heart stent in the divided image is improved, the range for detecting the heart stent is reduced, and the difficulty for detecting the heart stent is reduced.
Fig. 7 is a schematic flow chart of a method for detecting a cardiac stent according to another embodiment of the present application. The embodiment shown in fig. 7 is extended from the embodiment shown in fig. 6, and differences between the embodiment shown in fig. 7 and the embodiment shown in fig. 6 are described in detail, so that the description is omitted.
As shown in fig. 7, in the embodiment of the present application, the step of dividing the three-dimensional image of the chest region to obtain the three-dimensional image of the heart region includes the following steps.
And 710, performing dicing processing on the three-dimensional chest region image to obtain a plurality of three-dimensional chest image diced pieces.
Specifically, among the plurality of three-dimensional chest image cuts, there is an overlapping region between adjacent three-dimensional chest image cuts. The volume of the overlapping region may be half the volume of the three-dimensional chest image slice. The volume of the overlapping region may be other values, and the present application is not particularly limited.
In practical application, the dicing can be performed by a sliding window mode, for example, a fixed-volume dicing sliding window is selected, and a sliding window overlapping by 1/2 is continuously diced on the Z axis. The chest region three-dimensional image may be a chest region image sequence composed of a plurality of chest region images, and the Z-axis is a direction perpendicular to the plurality of chest region images.
At step 720, a three-dimensional image of the heart region is determined based on the plurality of three-dimensional chest image patches and the heart region detection model.
Specifically, the heart region detection model is used to detect heart regions in a plurality of three-dimensional chest image slices to determine a heart region three-dimensional image. The heart region detection model may be a network model.
The three-dimensional chest image is diced to obtain a plurality of three-dimensional chest image diced, and in the plurality of three-dimensional chest image diced, an overlapping area is arranged between adjacent three-dimensional chest image diced, so that the information of the three-dimensional chest image is prevented from being omitted in the dicing process, and the accuracy of the determined three-dimensional heart image is improved.
In one embodiment of the application, based on the plurality of three-dimensional chest image patches and the heart region detection model, determining the heart region three-dimensional image may be: and obtaining a plurality of detection areas based on the plurality of three-dimensional chest image dices and the heart area detection model, and then carrying out maximum connected domain processing on the plurality of detection areas so as to remove the heart false positive information and obtain a heart area three-dimensional image.
Fig. 8 is a schematic flow chart of a method for detecting a cardiac stent according to another embodiment of the present application. The embodiment shown in fig. 8 is extended from the embodiment shown in fig. 6, and differences between the embodiment shown in fig. 8 and the embodiment shown in fig. 6 are described in detail, so that details of the differences will not be repeated.
As shown in fig. 8, in the embodiment of the present application, the step of determining stent information in the coronary labeling three-dimensional image to be detected based on the coronary labeling three-dimensional image to be detected and the stent detection model includes the following steps.
Step 810, determining a plurality of coronary labeling cut blocks to be detected based on the coronary labeling three-dimensional image to be detected.
Step 820, determining stent region information corresponding to each of the coronary labeling dices to be detected based on the coronary labeling dices to be detected and the stent detection model.
Specifically, the stent information in the coronary artery labeling three-dimensional image to be detected comprises stent region information corresponding to each of the coronary artery labeling cut pieces to be detected.
In practical application, based on the coronary artery labeling three-dimensional image to be detected, a plurality of coronary artery labeling cut blocks to be detected are determined, and then based on the coronary artery labeling cut blocks to be detected and the stent detection model, stent region information corresponding to the coronary artery labeling cut blocks to be detected is determined, so that the requirement of executing the heart stent detection method on hardware is reduced, and the heart stent detection efficiency is improved.
In an embodiment of the present application, before determining a plurality of coronary labeling cut blocks to be detected based on the coronary labeling three-dimensional image to be detected, linear normalization processing and resampling processing may be performed on the coronary labeling three-dimensional image to be detected. By way of example only, and in an illustrative, the coronary labeling three-dimensional image to be detected can be resampled to 128 x 128. By performing linear normalization processing and resampling processing on the coronary artery labeling three-dimensional image to be detected, the data volume input into the stent detection model is reduced, and the detection efficiency of the cardiac stent is further improved.
Fig. 9 is a schematic flow chart of a method for detecting a cardiac stent according to another embodiment of the present application. The embodiment shown in fig. 9 is extended from the embodiment shown in fig. 8, and differences between the embodiment shown in fig. 9 and the embodiment shown in fig. 8 are described in detail, so that details of the differences will not be repeated.
As shown in fig. 9, in the embodiment of the present application, after the step of determining stent region information corresponding to each of the plurality of coronary artery labeling cuts to be detected based on the plurality of coronary artery labeling cuts to be detected and the stent detection model, the following steps are further included.
Step 910, mapping stent region information corresponding to each of the plurality of coronary labeling cuts to be detected to the three-dimensional image of the heart region, so as to determine a stent region in the three-dimensional image of the heart region.
Specifically, the stent region information corresponding to each of the plurality of coronary artery labeling cut pieces to be detected is mapped to the heart region three-dimensional image, the coronary artery labeling cut pieces to be detected including the stent region information can be resampled to be the same as the resolution of the heart region three-dimensional image, and then the resampled coronary artery labeling cut pieces to be detected are mapped to the heart region three-dimensional image.
In practical application, after a plurality of coronary artery labeling cut blocks to be detected are input into a stent detection model, the stent detection model outputs stent region information corresponding to the coronary artery labeling cut blocks to be detected. Therefore, the stent region information corresponding to each of the coronary artery labeling cut pieces to be detected is mapped to the heart region three-dimensional image, and the stent region in the heart region three-dimensional image can be determined. Namely, the stent region can be marked in the three-dimensional image of the heart region, which is convenient for a doctor to check.
In an embodiment of the present application, after mapping stent region information corresponding to each of a plurality of coronary artery labeling and cutting blocks to be detected to a three-dimensional image of a heart region, a plurality of stent regions can be obtained, and then, a maximum connected domain processing is performed on the plurality of stent regions, so that stent false positive is removed, and a stent region in the three-dimensional image of the heart region is obtained.
In step 920, the stent region information corresponding to each of the coronary labeling cut pieces to be detected is mapped to the chest region three-dimensional image, so as to determine the stent region in the chest region three-dimensional image.
Specifically, the stent region information corresponding to each of the plurality of coronary artery labeling cuts to be detected is mapped to the chest region three-dimensional image, the coronary artery labeling cuts to be detected including the stent region information can be resampled to be the same as the resolution of the chest region three-dimensional image, and then the resampled coronary artery labeling cuts to be detected are mapped to the chest region three-dimensional image.
In practical application, after a plurality of coronary artery labeling cut blocks to be detected are input into a stent detection model, the stent detection model outputs stent region information corresponding to the coronary artery labeling cut blocks to be detected. Therefore, the stent region information corresponding to each of the coronary artery labeling cut pieces to be detected is mapped to the chest region three-dimensional image, and the stent region in the chest region three-dimensional image can be determined. Namely, the bracket area can be marked in the three-dimensional image of the chest area, so that the doctor can conveniently check the bracket area.
In an embodiment of the present application, after mapping stent region information corresponding to each of a plurality of coronary artery labeling and cutting blocks to be detected to a chest region three-dimensional image, a plurality of stent regions may be obtained, and then, a maximum connected domain processing is performed on the plurality of stent regions, so as to remove stent false positive, and obtain a stent region in the chest region three-dimensional image.
Fig. 10 is a flowchart of a method for detecting a cardiac stent according to another embodiment of the present application. The embodiment shown in fig. 10 is extended from the embodiment shown in fig. 7, and differences between the embodiment shown in fig. 10 and the embodiment shown in fig. 7 are described in detail, so that the description is omitted.
As shown in fig. 10, in the embodiment of the present application, the step of determining a three-dimensional image of a heart region based on a plurality of three-dimensional chest image slices and a heart region detection model includes the following steps.
Step 1010, determining a heart region in the three-dimensional chest region image based on the plurality of three-dimensional chest image patches and the heart region detection model.
And 1020, intercepting a heart region in the three-dimensional image of the chest region by using an circumscribed cuboid algorithm to determine the three-dimensional image of the heart region.
In practical application, based on a plurality of three-dimensional chest image dices and a heart region detection model, determining a heart region in the chest region three-dimensional image, only displaying the heart region, then acquiring coordinates of an circumscribed cuboid of the heart region, and finally intercepting the heart region three-dimensional image by utilizing the coordinates of the circumscribed cuboid of the heart region. For example, the coordinates of the circumscribed cuboid of the heart region may be [ xmin, ymin, zmin, xmax, ymax, zmax ], i.e. the coordinates of the two ends of the diagonal of the circumscribed cuboid.
The heart region in the three-dimensional chest region image is determined based on a plurality of three-dimensional chest image cutting blocks and a heart region detection model, and then the heart region in the three-dimensional chest region image is intercepted by utilizing an external cuboid algorithm so as to determine the three-dimensional heart region image, so that the three-dimensional heart region image can be completely intercepted, and a complete and accurate data base is provided for subsequent detection of the heart support.
The method embodiment of the present application is described above in detail with reference to fig. 1 to 10, and the apparatus embodiment of the present application is described below in detail with reference to fig. 11 to 19. It is to be understood that the description of the method embodiments corresponds to the description of the device embodiments, and that parts not described in detail can therefore be seen in the preceding method embodiments.
Exemplary apparatus
Fig. 11 is a schematic structural diagram of a cardiac stent detection device according to an embodiment of the present application. As shown in fig. 11, a cardiac stent detection apparatus 1100 according to an embodiment of the present application includes: an acquisition module 1110 and a detection module 1120.
Specifically, the obtaining module 1110 is configured to obtain a coronary artery labeling three-dimensional image to be detected, where the coronary artery labeling three-dimensional image to be detected includes labeled coronary artery information. The detection module 1120 is configured to determine stent information in the coronary labeling three-dimensional image to be detected based on the coronary labeling three-dimensional image to be detected and a stent detection model, wherein the stent detection model is used for detecting the coronary labeling three-dimensional image to be detected to determine the stent information in the coronary labeling three-dimensional image to be detected.
Fig. 12 is a schematic structural diagram of a cardiac stent detection device according to another embodiment of the present application. The embodiment shown in fig. 12 is extended from the embodiment shown in fig. 11, and differences between the embodiment shown in fig. 12 and the embodiment shown in fig. 11 are described in detail, so that details of the differences will not be repeated.
As shown in fig. 12, the acquisition module 1110 of the embodiment of the present application includes: the first centerline determination unit 1111, the second centerline determination unit 1112, and the coronary labeling unit 1113.
Specifically, the first center line determination unit 1111 is configured to perform a center line extraction operation on the coronary information contained in the three-dimensional image of the heart region, resulting in a first coronary center line. The second centerline determination unit 1112 is configured to perform morphological dilation processing on the first coronary artery centerline to obtain a second coronary artery centerline. The coronary labeling unit 1113 is configured to label the three-dimensional image of the heart region based on the second coronary centerline, resulting in a coronary labeling three-dimensional image to be detected.
Fig. 13 is a schematic structural diagram of a cardiac stent detection device according to another embodiment of the present application. The embodiment shown in fig. 13 is extended from the embodiment shown in fig. 12, and differences between the embodiment shown in fig. 13 and the embodiment shown in fig. 12 are described in detail, so that details of the differences will not be repeated.
As shown in fig. 13, the coronary labeling unit 1113 according to the embodiment of the present application includes: a third centerline-determining subunit 1310 and a coronary labeling subunit 1320.
Specifically, the third centerline determination subunit 1310 is configured to perform gaussian blur processing on the second coronary artery centerline to obtain a third coronary artery centerline. The coronary labeling subunit 1320 is configured to label the three-dimensional image of the heart region based on the third coronary centerline, resulting in a coronary labeling three-dimensional image to be detected.
Fig. 14 is a schematic structural diagram of a cardiac stent detection device according to another embodiment of the present application. The embodiment shown in fig. 14 is extended from the embodiment shown in fig. 11, and differences between the embodiment shown in fig. 14 and the embodiment shown in fig. 11 are described in detail, so that details of the differences will not be repeated.
As shown in fig. 14, the cardiac stent detection apparatus 1100 according to the embodiment of the present application further includes: a construction module 1130 and a training module 1140.
Specifically, the construction module 1130 is configured to construct an initial network model and a loss function of the initial network model, wherein a loss value of the loss function is determined based on the coronary region information in the three-dimensional image sample during the training process. The training module 1140 is configured to train the initial network model based on the three-dimensional image sample, the coronary labeling information and the stent labeling information corresponding to the three-dimensional image sample, and the loss function to generate a stent detection model.
Fig. 15 is a schematic structural diagram of a cardiac stent detection device according to another embodiment of the present application. The embodiment shown in fig. 15 is extended from the embodiment shown in fig. 11, and differences between the embodiment shown in fig. 15 and the embodiment shown in fig. 11 are described in detail, so that details of the differences will not be repeated.
As shown in fig. 15, the cardiac stent detection apparatus 1100 according to the embodiment of the present application further includes: a segmentation module 1150.
Specifically, the segmentation module 1150 is configured to segment the chest region three-dimensional image to obtain a heart region three-dimensional image, where the heart region three-dimensional image is a portion of the chest region three-dimensional image.
Fig. 16 is a schematic structural diagram of a cardiac stent detection device according to another embodiment of the present application. The embodiment shown in fig. 16 is extended from the embodiment shown in fig. 15, and differences between the embodiment shown in fig. 16 and the embodiment shown in fig. 15 are described in detail, so that details of the differences will not be repeated.
As shown in fig. 16, the dividing module 1150 according to the embodiment of the present application includes: a dicing unit 1151 and a heart region determining unit 1152.
Specifically, the dicing unit 1151 is configured to perform dicing processing on the three-dimensional chest region image to obtain a plurality of three-dimensional chest image diced pieces, wherein among the plurality of three-dimensional chest image diced pieces, there is an overlapping region between adjacent three-dimensional chest image diced pieces. The heart region determination unit 1152 is configured to determine a heart region three-dimensional image based on a plurality of three-dimensional chest image patches and a heart region detection model for detecting a heart region in the plurality of three-dimensional chest image patches to determine the heart region three-dimensional image.
Fig. 17 is a schematic structural diagram of a cardiac stent detection device according to another embodiment of the present application. The embodiment shown in fig. 17 is extended from the embodiment shown in fig. 15, and differences between the embodiment shown in fig. 17 and the embodiment shown in fig. 15 are described in detail, so that details of the differences will not be repeated.
As shown in fig. 17, the detection module 1120 according to the embodiment of the present application includes: a coronary artery labeling cut determination unit 1121 and a stent region information determination unit 1122.
Specifically, the coronary artery labeling cut-out determination unit 1121 is configured to determine a plurality of coronary artery labeling cuts to be detected based on the coronary artery labeling three-dimensional image to be detected. The stent region information determining unit 1122 is configured to determine stent region information corresponding to each of the plurality of coronary labeling cuts to be detected based on the plurality of coronary labeling cuts to be detected and the stent detection model, wherein the stent information in the coronary labeling three-dimensional image to be detected includes the stent region information corresponding to each of the plurality of coronary labeling cuts to be detected.
Fig. 18 is a schematic structural diagram of a cardiac stent detection device according to another embodiment of the present application. The embodiment shown in fig. 18 is extended from the embodiment shown in fig. 17, and differences between the embodiment shown in fig. 18 and the embodiment shown in fig. 17 are described in detail, so that details of the differences will not be repeated.
As shown in fig. 18, the cardiac stent detection apparatus 1100 according to the embodiment of the present application further includes: a first mapping module 1160.
Specifically, the first mapping module 1160 is configured to map stent region information corresponding to each of the plurality of coronary labeling cuts to be detected to a three-dimensional image of a heart region to determine a stent region in the three-dimensional image of the heart region.
As shown in fig. 18, the cardiac stent detection apparatus 1100 according to the embodiment of the present application further includes: the second mapping module 1170.
Specifically, the second mapping module 1170 is configured to map stent region information corresponding to each of the plurality of coronary labeling cuts to be detected to the chest region three-dimensional image to determine a stent region in the chest region three-dimensional image.
Fig. 19 is a schematic structural diagram of a cardiac stent detection device according to another embodiment of the present application. The embodiment shown in fig. 19 is extended from the embodiment shown in fig. 16, and differences between the embodiment shown in fig. 19 and the embodiment shown in fig. 16 are described in detail, so that the description is omitted.
As shown in fig. 19, the heart region determining unit 1152 of the embodiment of the present application includes: a heart region determination subunit 1910 and an interception subunit 1920.
Specifically, the heart region determination subunit 1910 is configured to determine a heart region in the chest region three-dimensional image based on the plurality of three-dimensional chest image patches and the heart region detection model. The clipping sub-unit 1920 is configured to clip a heart region in the three-dimensional image of the chest region using an circumscribed cuboid algorithm to determine a three-dimensional image of the heart region.
The operations and functions of the acquisition module 1110, the detection module 1120, the construction module 1130, the training module 1140, the segmentation module 1150, the first mapping module 1160, and the second mapping module 1170 in the cardiac stent detection device provided in fig. 11 to 19, and the first center line determination unit 1111, the second center line determination unit 1112, and the coronary labeling unit 1113 included in the acquisition module 1110, and the third center line determination unit 1310 and the coronary labeling unit 1320 included in the coronary labeling unit 1113, and the dicing unit 1151 and the heart region determination unit 1152 included in the segmentation module 1150, and the coronary labeling dicing determination unit 1121 and the stent region information determination unit 1122 included in the detection module 1120, and the heart region determination unit 1910 and the clipping unit 1920 included in the heart region determination unit 1152 may refer to the cardiac stent detection method provided in fig. 2 to 10 described above, and will not be repeated here.
Exemplary electronic device
Fig. 20 is a schematic structural diagram of an electronic device according to an embodiment of the application. As shown in fig. 20, the electronic device 2000 includes: one or more processors 2001 and memory 2002; and computer program instructions stored in the memory 2002, which when executed by the processor 2001, cause the processor 2001 to perform the cardiac stent detection method of any of the embodiments described above.
The processor 2001 may be a central processing unit (Central Processing Unit, CPU) or other form of processing unit having data transmission capabilities and/or instruction execution capabilities, and may control other components in the electronic device to perform desired functions.
Memory 2002 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory. Volatile memory can include, for example, random access memory (Random Access Memory, RAM) and/or Cache memory (Cache), among others. The nonvolatile Memory may include, for example, a Read Only Memory (ROM), a hard disk, a flash Memory, and the like. One or more computer program instructions may be stored on a computer readable storage medium and the processor 2001 may execute the program instructions to implement the steps in the cardiac stent detection method of the various embodiments of the present application above and/or other desired functions.
In one example, the electronic device 2000 may further include: an input device 2003 and an output device 2004, which are interconnected by a bus system and/or other forms of connection mechanisms (not shown in fig. 20).
In addition, the input device 2003 may also include, for example, a keyboard, a mouse, a microphone, and the like.
The output device 2004 can output various information to the outside. The output means 2004 may include, for example, a display, speakers, a printer, and a communication network and remote output devices connected thereto, etc.
Of course, only some of the components of the electronic device 2000 relevant to the present application are shown in fig. 20 for simplicity, and components such as buses, input devices/output interfaces, etc. are omitted. In addition, the electronic device 2000 may include any other suitable components depending on the particular application.
Exemplary computer-readable storage Medium
In addition to the methods and apparatus described above, embodiments of the application may also be a computer program product comprising computer program instructions which, when executed by a processor, cause the processor to perform the steps of the cardiac stent detection method as described in any of the embodiments described above.
The computer program product may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device, partly on a remote computing device, or entirely on the remote computing device or server.
Furthermore, embodiments of the application may also be a computer-readable storage medium, on which computer program instructions are stored which, when being executed by a processor, cause the processor to perform the steps in a cardiac stent detection method according to various embodiments of the application described in the "exemplary methods" section of the description above.
A computer readable storage medium may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may include, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium would include the following: an electrical connection having one or more wires, a portable disk, a hard disk, RAM, ROM, erasable programmable read-only memory (Erasable Programmable Read Only Memory, EPROM) or flash memory, optical fiber, portable compact disk read-only memory (Compact Disk Read Only Memory, CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
The basic principles of the present application have been described above in connection with specific embodiments, however, it should be noted that the advantages, benefits, effects, etc. mentioned in the present application are merely examples and not intended to be limiting, and these advantages, benefits, effects, etc. are not to be considered as essential to the various embodiments of the present application. Furthermore, the specific details disclosed herein are for purposes of illustration and understanding only, and are not intended to be limiting, as the application is not necessarily limited to practice with the above described specific details.
The block diagrams of the devices, apparatuses, devices, systems referred to in the present application are only illustrative examples and are not intended to require or imply that the connections, arrangements, configurations must be made in the manner shown in the block diagrams. As will be appreciated by one of skill in the art, the devices, apparatuses, devices, systems may be connected, arranged, configured in any manner. Words such as "including," "comprising," "having," and the like are words of openness and mean "including but not limited to," and are used interchangeably therewith. The terms "or" and "as used herein refer to and are used interchangeably with the term" and/or "unless the context clearly indicates otherwise. The term "such as" as used herein refers to, and is used interchangeably with, the phrase "such as, but not limited to.
It is also noted that in the apparatus, devices and methods of the present application, the components or steps may be disassembled and/or assembled. Such decomposition and/or recombination should be considered as equivalent aspects of the present application.
The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit embodiments of the application to the form disclosed herein. Although a number of example aspects and embodiments have been discussed above, a person of ordinary skill in the art will recognize certain variations, modifications, alterations, additions, and subcombinations thereof.
The foregoing is only illustrative of the present application and is not to be construed as limiting thereof, but rather as presently claimed, and is intended to cover all modifications, alternatives, and equivalents falling within the spirit and scope of the application.

Claims (11)

1. A method for detecting a cardiac stent, comprising:
performing central line extraction operation on coronary artery information contained in the three-dimensional image of the heart region to obtain a first coronary artery central line;
performing morphological expansion processing on the first coronary artery central line to obtain a second coronary artery central line, wherein the second coronary artery central line is used for hiding real width information of the coronary artery;
labeling the heart region three-dimensional image based on the second coronary artery central line to obtain a coronary artery labeling three-dimensional image to be detected, wherein the coronary artery labeling three-dimensional image to be detected contains labeled coronary artery information;
And determining stent information in the coronary artery labeling three-dimensional image to be detected based on the coronary artery labeling three-dimensional image to be detected and a stent detection model, wherein the stent detection model is used for detecting the coronary artery labeling three-dimensional image to be detected so as to determine the stent information in the coronary artery labeling three-dimensional image to be detected.
2. The method for detecting a cardiac stent according to claim 1, wherein the labeling the three-dimensional image of the heart region based on the second coronary artery centerline to obtain the coronary artery labeling three-dimensional image to be detected includes:
performing Gaussian blur processing on the second coronary artery central line to obtain a third coronary artery central line;
and labeling the heart region three-dimensional image based on the third coronary artery central line to obtain the coronary artery labeling three-dimensional image to be detected.
3. The method for detecting a cardiac stent according to claim 1, further comprising, before the center line extraction operation is performed on the coronary information included in the three-dimensional image of the heart region to obtain the first coronary center line:
and dividing the chest region three-dimensional image to obtain the heart region three-dimensional image, wherein the heart region three-dimensional image is a part of the chest region three-dimensional image.
4. A method of detecting a cardiac stent as defined in claim 3, wherein the segmenting the three-dimensional image of the chest region to obtain the three-dimensional image of the heart region comprises:
the three-dimensional chest image of the chest area is diced to obtain a plurality of three-dimensional chest image diced pieces, wherein an overlapping area is arranged between the adjacent three-dimensional chest image diced pieces in the plurality of three-dimensional chest image diced pieces;
determining the heart region three-dimensional image based on the plurality of three-dimensional chest image patches and a heart region detection model, wherein the heart region detection model is used for detecting heart regions in the plurality of three-dimensional chest image patches to determine the heart region three-dimensional image.
5. The method for detecting a cardiac stent according to claim 3, wherein the determining stent information in the coronary labeling three-dimensional image to be detected based on the coronary labeling three-dimensional image to be detected and a stent detection model includes:
determining a plurality of coronary artery labeling cut blocks to be detected based on the coronary artery labeling three-dimensional image to be detected;
and determining stent region information corresponding to each of the coronary artery labeling cut pieces to be detected based on the coronary artery labeling cut pieces to be detected and the stent detection model, wherein the stent information in the coronary artery labeling three-dimensional image to be detected comprises the stent region information corresponding to each of the coronary artery labeling cut pieces to be detected.
6. The method according to claim 5, further comprising, after the determining stent region information corresponding to each of the plurality of coronary labeling cuts to be detected based on the plurality of coronary labeling cuts to be detected and the stent detection model:
mapping stent region information corresponding to each of the coronary artery labeling cut pieces to be detected to the heart region three-dimensional image so as to determine a stent region in the heart region three-dimensional image; or (b)
And mapping the stent region information corresponding to each of the coronary artery labeling cut pieces to be detected to the chest region three-dimensional image so as to determine the stent region in the chest region three-dimensional image.
7. The method of claim 4, wherein the determining the three-dimensional image of the heart region based on the plurality of three-dimensional chest image slices and a heart region detection model comprises:
determining a heart region in the chest region three-dimensional image based on the plurality of three-dimensional chest image patches and a heart region detection model;
and intercepting a heart region in the chest region three-dimensional image by using an circumscribed cuboid algorithm so as to determine the heart region three-dimensional image.
8. The cardiac stent detection method according to any one of claims 1 to 7, further comprising, before the determining stent information in the coronary labeling three-dimensional image to be detected based on the coronary labeling three-dimensional image to be detected and a stent detection model:
constructing an initial network model and a loss function of the initial network model, wherein in the training process, the loss value of the loss function is determined based on coronary region information in a three-dimensional image sample;
and training the initial network model based on the three-dimensional image sample, coronary artery labeling information and stent labeling information corresponding to the three-dimensional image sample and the loss function to generate the stent detection model.
9. A cardiac stent detection device, comprising:
the acquisition module is configured to perform central line extraction operation on coronary artery information contained in the three-dimensional image of the heart region to obtain a first coronary artery central line; performing morphological expansion processing on the first coronary artery central line to obtain a second coronary artery central line, wherein the second coronary artery central line is used for hiding real width information of the coronary artery; labeling the heart region three-dimensional image based on the second coronary artery central line to obtain a coronary artery labeling three-dimensional image to be detected, wherein the coronary artery labeling three-dimensional image to be detected contains labeled coronary artery information;
The detection module is configured to determine stent information in the coronary artery labeling three-dimensional image to be detected based on the coronary artery labeling three-dimensional image to be detected and a stent detection model, wherein the stent detection model is used for detecting the coronary artery labeling three-dimensional image to be detected so as to determine the stent information in the coronary artery labeling three-dimensional image to be detected.
10. A computer readable storage medium, characterized in that the storage medium stores instructions that, when executed by a processor of an electronic device, enable the electronic device to perform the cardiac stent detection method of any one of the preceding claims 1 to 8.
11. An electronic device, the electronic device comprising:
a processor;
a memory for storing computer-executable instructions;
the processor for executing the computer-executable instructions to implement the cardiac stent detection method of any one of the preceding claims 1 to 8.
CN202210429123.1A 2022-04-22 2022-04-22 Heart stent detection method and device, storage medium and electronic equipment Active CN114782366B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210429123.1A CN114782366B (en) 2022-04-22 2022-04-22 Heart stent detection method and device, storage medium and electronic equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210429123.1A CN114782366B (en) 2022-04-22 2022-04-22 Heart stent detection method and device, storage medium and electronic equipment

Publications (2)

Publication Number Publication Date
CN114782366A CN114782366A (en) 2022-07-22
CN114782366B true CN114782366B (en) 2023-08-22

Family

ID=82430292

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210429123.1A Active CN114782366B (en) 2022-04-22 2022-04-22 Heart stent detection method and device, storage medium and electronic equipment

Country Status (1)

Country Link
CN (1) CN114782366B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108335284A (en) * 2018-01-09 2018-07-27 北京理工大学 A kind of coronary artery vessel centerline matching process and system
CN112967279A (en) * 2021-04-02 2021-06-15 慧影医疗科技（北京）有限公司 Method, device, storage medium and electronic equipment for detecting pulmonary nodules
CN113222964A (en) * 2021-05-27 2021-08-06 推想医疗科技股份有限公司 Method and device for generating coronary artery central line extraction model
CN113313715A (en) * 2021-05-27 2021-08-27 推想医疗科技股份有限公司 Method, device, apparatus and medium for segmenting cardiac artery blood vessel
CN113397579A (en) * 2021-07-23 2021-09-17 上海友脉科技有限责任公司 Hemodynamics analysis device, method, medium, and electronic device

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8855396B2 (en) * 2010-05-25 2014-10-07 Siemens Medical Solutions Usa, Inc. System for detecting an invasive anatomical instrument
US20120002855A1 (en) * 2010-06-30 2012-01-05 Fujifilm Corporation Stent localization in 3d cardiac images
US10699407B2 (en) * 2018-04-11 2020-06-30 Pie Medical Imaging B.V. Method and system for assessing vessel obstruction based on machine learning

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108335284A (en) * 2018-01-09 2018-07-27 北京理工大学 A kind of coronary artery vessel centerline matching process and system
CN112967279A (en) * 2021-04-02 2021-06-15 慧影医疗科技（北京）有限公司 Method, device, storage medium and electronic equipment for detecting pulmonary nodules
CN113222964A (en) * 2021-05-27 2021-08-06 推想医疗科技股份有限公司 Method and device for generating coronary artery central line extraction model
CN113313715A (en) * 2021-05-27 2021-08-27 推想医疗科技股份有限公司 Method, device, apparatus and medium for segmenting cardiac artery blood vessel
CN113397579A (en) * 2021-07-23 2021-09-17 上海友脉科技有限责任公司 Hemodynamics analysis device, method, medium, and electronic device

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
基于光学相干断层成像含支架的冠脉血管三维重建;仓定举;王笑梅;郑亚州;;上海师范大学学报(自然科学版)(第04期);全文 *

Also Published As

Publication number Publication date
CN114782366A (en) 2022-07-22

Similar Documents

Publication Publication Date Title
CN102460471B (en) Systems for computer aided lung nodule detection in chest tomosynthesis imaging
EP1469778B1 (en) Method and apparatus for semi-automatic aneurysm measurement and stent planning using volume image data
US7742639B2 (en) Data set visualization
CN102422307A (en) Method, system, apparatus, and computer program product for interactive hepatic vascular and biliary system assessment
US8768436B2 (en) Coronary artery angiography image processing method to detect occlusion and degree effect of blood vessel occlusion to an organ
CN111815598B (en) Calculation method, device and equipment of vascular parameters and storage medium
JPH07143977A (en) Digital image processing method for automatic detection of constriction
CN112862833A (en) Blood vessel segmentation method, electronic device and storage medium
CN112446866A (en) Blood flow parameter calculation method, device, equipment and storage medium
CN117115150B (en) Method, computing device and medium for determining branch vessels
CN111145160A (en) Method, device, server and medium for determining coronary artery branch where calcified area is located
CN112419484A (en) Three-dimensional blood vessel synthesis method and system, coronary artery analysis system and storage medium
CN112446867A (en) Method, device and equipment for determining blood flow parameters and storage medium
CN112132882A (en) Method and device for extracting blood vessel central line from coronary artery two-dimensional contrast image
EP3971830A1 (en) Pneumonia sign segmentation method and apparatus, medium and electronic device
CN114782366B (en) Heart stent detection method and device, storage medium and electronic equipment
CN111192664A (en) Image marking method, system, equipment and storage medium
US20230222668A1 (en) Image processing apparatus, image processing method, and recording medium
CN116050691A (en) Medical image report evaluation method, device, electronic equipment and storage medium
CN112137693B (en) Imaging method and device for four-dimensional ultrasonic guided puncture
CN111798468B (en) Image processing method and device, storage medium and electronic terminal
CN115761228A (en) Coronary artery calcified plaque segmentation method, device, equipment and storage medium
CN115330696A (en) Detection method, device and equipment of bracket and storage medium
Martinke et al. Bone Fracture and Lesion Assessment using Shape-Adaptive Unfolding.
CN114880960A (en) Method for evaluating radioactive embolism dose and injection position based on fluid dynamics

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant