CN116862874A - Blood vessel contour extraction method and device, electronic equipment and storage medium - Google Patents
Blood vessel contour extraction method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN116862874A CN116862874A CN202310837553.1A CN202310837553A CN116862874A CN 116862874 A CN116862874 A CN 116862874A CN 202310837553 A CN202310837553 A CN 202310837553A CN 116862874 A CN116862874 A CN 116862874A
- Authority
- CN
- China
- Prior art keywords
- blood vessel
- coordinate point
- point set
- loss function
- function value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 210000004204 blood vessel Anatomy 0.000 title claims abstract description 344
- 238000000605 extraction Methods 0.000 title claims description 42
- 238000010586 diagram Methods 0.000 claims abstract description 29
- 238000000034 method Methods 0.000 claims abstract description 25
- 238000006243 chemical reaction Methods 0.000 claims abstract description 18
- 230000006870 function Effects 0.000 claims description 123
- 238000012545 processing Methods 0.000 claims description 38
- 238000013528 artificial neural network Methods 0.000 claims description 16
- 238000012549 training Methods 0.000 claims description 15
- 238000002372 labelling Methods 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 5
- 239000000523 sample Substances 0.000 description 87
- 238000002608 intravascular ultrasound Methods 0.000 description 12
- 239000013598 vector Substances 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 7
- 238000002604 ultrasonography Methods 0.000 description 6
- 210000004351 coronary vessel Anatomy 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 238000002059 diagnostic imaging Methods 0.000 description 4
- 238000001990 intravenous administration Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 208000024172 Cardiovascular disease Diseases 0.000 description 2
- 208000031481 Pathologic Constriction Diseases 0.000 description 2
- 238000010009 beating Methods 0.000 description 2
- 208000026106 cerebrovascular disease Diseases 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 238000003708 edge detection Methods 0.000 description 2
- 230000002526 effect on cardiovascular system Effects 0.000 description 2
- 230000036541 health Effects 0.000 description 2
- 230000003902 lesion Effects 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 230000036262 stenosis Effects 0.000 description 2
- 208000037804 stenosis Diseases 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000002583 angiography Methods 0.000 description 1
- 239000003795 chemical substances by application Substances 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/149—Segmentation; Edge detection involving deformable models, e.g. active contour models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/42—Global feature extraction by analysis of the whole pattern, e.g. using frequency domain transformations or autocorrelation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20112—Image segmentation details
- G06T2207/20116—Active contour; Active surface; Snakes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30101—Blood vessel; Artery; Vein; Vascular
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Quality & Reliability (AREA)
- Radiology & Medical Imaging (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Ultra Sonic Daignosis Equipment (AREA)
Abstract
The application provides a method and a device for extracting a blood vessel contour, electronic equipment and a storage medium, wherein the method comprises the following steps: inputting the acquired blood vessel cross-section image set of the 3D blood vessel image to be identified into a pre-trained contour point identification model, and determining a first target polar coordinate point set of the inner diameter of the blood vessel and a second target polar coordinate point set of the outer diameter of the blood vessel in each blood vessel cross-section image to be identified; respectively carrying out coordinate conversion treatment on a first target polar coordinate point set and a second target polar coordinate point set of each to-be-identified blood vessel cross-section diagram, and determining a first target rectangular coordinate point set of the inner diameter of the blood vessel and a second target rectangular coordinate point set of the outer diameter of the blood vessel in each to-be-identified blood vessel cross-section diagram; and sequentially carrying out curve fitting treatment on each target rectangular coordinate point set to determine a blood vessel inner diameter contour curve and a blood vessel outer diameter contour curve of the 3D blood vessel image to be identified. Therefore, through the technical scheme of the application, the speed and accuracy of extracting the inner and outer diameter contours of the blood vessel can be effectively improved.
Description
Technical Field
The present application relates to the field of medical image processing technologies, and in particular, to a method and apparatus for extracting a blood vessel contour, an electronic device, and a storage medium.
Background
Worldwide, cardiovascular and cerebrovascular diseases have become one of the main diseases threatening human health, so in medical imaging, accurate segmentation of blood vessels such as coronary vessels and extraction of vessel contours are of great importance for determining stenosis, lesions, etc. of coronary arteries.
In the prior art, intravascular ultrasound (intravenous ultrasound, IVUS) technology is generally used for intravascular-intimal edge detection, wherein intravascular ultrasound (intravenous ultrasound, IVUS) refers to a medical imaging technology that uses a special catheter with an ultrasound probe connected to its end, in combination with an invasive catheter technology.
However, since the IVUS image twists with the beating of the heart, the extraction of the lumen intima profile is not favored, resulting in a slower and less accurate extraction of the lumen intima profile of the vessel. Therefore, how to improve the accuracy of the vessel contour extraction is a technical problem that needs to be solved by those skilled in the art.
Disclosure of Invention
Accordingly, the present application is directed to a method, an apparatus, an electronic device, and a storage medium for extracting a blood vessel contour, which can effectively improve the speed and accuracy of extracting an inner and outer diameter contour of a blood vessel.
The embodiment of the application provides a method for extracting a blood vessel contour, which comprises the following steps:
acquiring a blood vessel section image set of a 3D blood vessel image to be identified; the blood vessel section image set comprises a plurality of blood vessel section images to be identified, wherein the blood vessel section images to be identified are images intercepted according to a central line point on a blood vessel central line of the 3D blood vessel image to be identified;
inputting the blood vessel section image set into a pre-trained contour point identification model, and determining a first target polar coordinate point set of the inner diameter of a blood vessel and a second target polar coordinate point set of the outer diameter of the blood vessel in each blood vessel section image set to be identified;
respectively carrying out coordinate conversion treatment on a first target polar coordinate point set of the inner diameter of the blood vessel and a second target polar coordinate point set of the outer diameter of the blood vessel in each blood vessel section to be identified, and determining a first target rectangular coordinate point set of the inner diameter of the blood vessel and a second target rectangular coordinate point set of the outer diameter of the blood vessel in each blood vessel section to be identified;
and sequentially performing curve fitting treatment on a first target rectangular coordinate point set of the inner diameter of the blood vessel and a second target rectangular coordinate point set of the outer diameter of the blood vessel in each to-be-identified blood vessel section view, and determining a contour curve of the inner diameter of the blood vessel and a contour curve of the outer diameter of the blood vessel of the to-be-identified 3D blood vessel image.
Optionally, the contour point identification model is constructed by:
acquiring a sample vessel cross-section image set of a plurality of sample 3D vessel images; wherein each sample vessel cross-sectional view in the sample vessel cross-sectional image set corresponds to a first set of real polar points having an inner vessel diameter and a second set of real polar points having an outer vessel diameter;
inputting a sample vessel section image set of each sample 3D vessel image into an initial contour point identification neural network in sequence, and predicting a first predicted polar coordinate point set of the inner diameter of a vessel and a second predicted polar coordinate point set of the outer diameter of the vessel in each sample vessel section image;
determining a target loss function value based on the first real polar coordinate point set, the second real polar coordinate point set, the first predicted polar coordinate point set and the second predicted polar coordinate point set of each sample vessel cross-sectional view;
and carrying out iterative training on the initial contour point recognition neural network based on the objective loss function value, and updating network parameters of the initial contour point recognition neural network until the objective loss function value is converged, and stopping training to obtain the contour point recognition model.
Optionally, determining a first set of real polar points for the inside diameter of the vessel and a second set of real polar points for the outside diameter of the vessel for each of the sample vessel cross-sectional images in the set of sample vessel cross-sectional images by:
Performing central line extraction processing on the sample 3D blood vessel image, and determining a blood vessel central line of the sample 3D blood vessel image; wherein the vessel centerline comprises a plurality of centerline points thereon;
for each centerline point, intercepting a plane with a preset size in the tangential direction of the centerline of the perpendicular blood vessel comprising the centerline point, and determining a sample blood vessel section diagram corresponding to the centerline point;
for each sample blood vessel section, based on the internal and external diameter contour labeling result on the sample blood vessel section, performing contour coordinate point extraction processing to determine a first real rectangular coordinate point set of the internal diameter of the blood vessel and a second real rectangular coordinate point set of the external diameter of the blood vessel in the sample blood vessel section;
and carrying out coordinate conversion processing on the basis of the central line points in the section of each sample blood vessel according to the real polar coordinate point set in the section of each sample blood vessel, and determining a first real polar coordinate point set of the inner diameter of the blood vessel and a second real polar coordinate point set of the outer diameter of the blood vessel in the section of each sample blood vessel.
Optionally, the determining the objective loss function value based on the first real polar coordinate point set, the second real polar coordinate point set, the first predicted polar coordinate point set, and the second predicted polar coordinate point set of each sample vessel cross-sectional view includes:
Determining a real interval set of the inner diameter of the blood vessel and the outer diameter of the blood vessel based on the first real polar coordinate point set and the second real polar coordinate point set;
determining a predicted distance set of the inner diameter of the blood vessel and the outer diameter of the blood vessel based on the first predicted polar coordinate point set and the second predicted polar coordinate point set;
comparing the first predicted polar coordinate point set with the first real polar coordinate point set to determine a first loss function value;
comparing the second predicted polar coordinate point set with the second real polar coordinate point set to determine a second loss function value;
comparing the predicted distance set with the real distance set to determine a third loss function value;
the objective loss function value is determined based on the first loss function value, the second loss function value, and the third loss function value.
Optionally, the determining the objective loss function value based on the first loss function value, the second loss function value, and the third loss function value includes:
and multiplying the first loss function value, the second loss function value and the third loss function value by the weight coefficients corresponding to the first loss function value, the second loss function value and the third loss function value respectively, and adding the three loss function values multiplied by the weight coefficients to determine the target loss function value.
Optionally, before performing curve fitting processing on the first target rectangular coordinate point set of the inner diameter of the blood vessel and the second target rectangular coordinate point set of the outer diameter of the blood vessel in each to-be-identified blood vessel section view in sequence, the extraction method further includes:
and respectively carrying out interpolation processing on the first target rectangular coordinate point set and the second target rectangular coordinate point set, and increasing the number of coordinate points in the first target rectangular coordinate point set and the second target rectangular coordinate point set.
Optionally, the contour point recognition model is a single-task recognition model or a multi-task recognition model.
The embodiment of the application also provides an extraction device of the blood vessel outline, which comprises:
the acquisition module is used for acquiring a blood vessel section image set of the 3D blood vessel image to be identified; the blood vessel section image set comprises a plurality of blood vessel section images to be identified, wherein the blood vessel section images to be identified are images intercepted according to a central line point on a blood vessel central line of the 3D blood vessel image to be identified;
the first determining module is used for inputting the blood vessel section image set into a pre-trained contour point identification model and determining a first target polar coordinate point set of the inner diameter of a blood vessel and a second target polar coordinate point set of the outer diameter of the blood vessel in each blood vessel section image set to be identified;
The coordinate conversion module is used for carrying out coordinate conversion treatment on a first target polar coordinate point set of the inner diameter of the blood vessel and a second target polar coordinate point set of the outer diameter of the blood vessel in each blood vessel section diagram to be identified respectively, and determining a first target rectangular coordinate point set of the inner diameter of the blood vessel and a second target rectangular coordinate point set of the outer diameter of the blood vessel in each blood vessel section diagram to be identified;
and the fitting processing module is used for sequentially carrying out curve fitting processing on a first target rectangular coordinate point set of the inner diameter of the blood vessel and a second target rectangular coordinate point set of the outer diameter of the blood vessel in each to-be-identified blood vessel section diagram, and determining a contour curve of the inner diameter of the blood vessel and a contour curve of the outer diameter of the blood vessel of the to-be-identified 3D blood vessel image.
Optionally, the extracting device further includes a model building module, where the model building module is configured to:
acquiring a sample vessel cross-section image set of a plurality of sample 3D vessel images; wherein each sample vessel cross-sectional view in the sample vessel cross-sectional image set corresponds to a first set of real polar points having an inner vessel diameter and a second set of real polar points having an outer vessel diameter;
inputting a sample vessel section image set of each sample 3D vessel image into an initial contour point identification neural network in sequence, and predicting a first predicted polar coordinate point set of the inner diameter of a vessel and a second predicted polar coordinate point set of the outer diameter of the vessel in each sample vessel section image;
Determining a target loss function value based on the first real polar coordinate point set, the second real polar coordinate point set, the first predicted polar coordinate point set and the second predicted polar coordinate point set of each sample vessel cross-sectional view;
and carrying out iterative training on the initial contour point recognition neural network based on the objective loss function value, and updating network parameters of the initial contour point recognition neural network until the objective loss function value is converged, and stopping training to obtain the contour point recognition model.
Optionally, the extracting device further includes a second determining module, where the second determining module is configured to:
performing central line extraction processing on the sample 3D blood vessel image, and determining a blood vessel central line of the sample 3D blood vessel image; wherein the vessel centerline comprises a plurality of centerline points thereon;
for each centerline point, intercepting a plane with a preset size in the tangential direction of the centerline of the perpendicular blood vessel comprising the centerline point, and determining a sample blood vessel section diagram corresponding to the centerline point;
for each sample blood vessel section, based on the internal and external diameter contour labeling result on the sample blood vessel section, performing contour coordinate point extraction processing to determine a first real rectangular coordinate point set of the internal diameter of the blood vessel and a second real rectangular coordinate point set of the external diameter of the blood vessel in the sample blood vessel section;
And carrying out coordinate conversion processing on the basis of the central line points in the section of each sample blood vessel according to the real polar coordinate point set in the section of each sample blood vessel, and determining a first real polar coordinate point set of the inner diameter of the blood vessel and a second real polar coordinate point set of the outer diameter of the blood vessel in the section of each sample blood vessel.
Optionally, the model building module is configured to, when determining the objective loss function value based on the first real polar coordinate point set, the second real polar coordinate point set, the first predicted polar coordinate point set, and the second predicted polar coordinate point set of each sample vessel cross-sectional view,:
determining a real interval set of the inner diameter of the blood vessel and the outer diameter of the blood vessel based on the first real polar coordinate point set and the second real polar coordinate point set;
determining a predicted distance set of the inner diameter of the blood vessel and the outer diameter of the blood vessel based on the first predicted polar coordinate point set and the second predicted polar coordinate point set;
comparing the first predicted polar coordinate point set with the first real polar coordinate point set to determine a first loss function value;
comparing the second predicted polar coordinate point set with the second real polar coordinate point set to determine a second loss function value;
Comparing the predicted distance set with the real distance set to determine a third loss function value;
the objective loss function value is determined based on the first loss function value, the second loss function value, and the third loss function value.
Optionally, the model building module, when configured to determine the objective loss function value based on the first loss function value, the second loss function value, and the third loss function value, is configured to:
and multiplying the first loss function value, the second loss function value and the third loss function value by the weight coefficients corresponding to the first loss function value, the second loss function value and the third loss function value respectively, and adding the three loss function values multiplied by the weight coefficients to determine the target loss function value.
Optionally, the extracting device further includes an interpolation module, where the interpolation module is configured to: before curve fitting processing is sequentially carried out on a first target rectangular coordinate point set of the inner diameter of a blood vessel and a second target rectangular coordinate point set of the outer diameter of the blood vessel in each blood vessel section diagram to be identified, interpolation processing is respectively carried out on the first target rectangular coordinate point set and the second target rectangular coordinate point set, and the number of coordinate points in the first target rectangular coordinate point set and the second target rectangular coordinate point set is increased.
Optionally, the contour point recognition model is a single-task recognition model or a multi-task recognition model.
The embodiment of the application also provides electronic equipment, which comprises: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, the processor and the memory in communication via the bus when the electronic device is running, the machine-readable instructions when executed by the processor performing the steps of the extraction method as described above.
The embodiments of the present application also provide a computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the extraction method as described above.
The embodiment of the application provides a method, a device, electronic equipment and a storage medium for extracting a blood vessel contour, which comprise the following steps: acquiring a blood vessel section image set of a 3D blood vessel image to be identified; the blood vessel section image set comprises a plurality of blood vessel section images to be identified, wherein the blood vessel section images to be identified are images intercepted according to a central line point on a blood vessel central line of the 3D blood vessel image to be identified; inputting the blood vessel section image set into a pre-trained contour point identification model, and determining a first target polar coordinate point set of the inner diameter of a blood vessel and a second target polar coordinate point set of the outer diameter of the blood vessel in each blood vessel section image set to be identified; respectively carrying out coordinate conversion treatment on a first target polar coordinate point set of the inner diameter of the blood vessel and a second target polar coordinate point set of the outer diameter of the blood vessel in each blood vessel section to be identified, and determining a first target rectangular coordinate point set of the inner diameter of the blood vessel and a second target rectangular coordinate point set of the outer diameter of the blood vessel in each blood vessel section to be identified; and sequentially performing curve fitting treatment on a first target rectangular coordinate point set of the inner diameter of the blood vessel and a second target rectangular coordinate point set of the outer diameter of the blood vessel in each to-be-identified blood vessel section view, and determining a contour curve of the inner diameter of the blood vessel and a contour curve of the outer diameter of the blood vessel of the to-be-identified 3D blood vessel image.
Therefore, by converting rectangular coordinates into polar coordinates, background interference is reduced, model learning is simplified, and model learning speed is improved; the loss function is calculated through the distance between the inner diameter and the outer diameter, so that the study of the thickness of the blood vessel wall is enhanced, the model identification precision is improved, and the accuracy of extracting the inner diameter profile and the outer diameter profile of the blood vessel is further improved.
In order to make the above objects, features and advantages of the present application more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a method for extracting a blood vessel contour according to an embodiment of the present application;
FIG. 2 is a schematic view of a blood vessel centerline according to the present application;
FIG. 3 is a schematic structural diagram of a labeling result of the outline of the inner diameter and the outer diameter of a blood vessel;
FIG. 4 is a schematic diagram of a coordinate transformation principle according to the present application;
FIG. 5 is a schematic view of the structure of an inner diameter contour line of a blood vessel according to the present application;
fig. 6 is a schematic structural diagram of a blood vessel contour extraction device according to an embodiment of the present application;
FIG. 7 is a second schematic diagram of a device for extracting a blood vessel contour according to an embodiment of the present application;
fig. 8 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. The components of the embodiments of the present application generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the application, as presented in the figures, is not intended to limit the scope of the application, as claimed, but is merely representative of selected embodiments of the application. Based on the embodiments of the present application, every other embodiment obtained by a person skilled in the art without making any inventive effort falls within the scope of protection of the present application.
Worldwide, cardiovascular and cerebrovascular diseases have become one of the main diseases threatening human health, so in medical imaging, accurate segmentation of blood vessels such as coronary vessels and extraction of vessel contours are of great importance for determining stenosis, lesions, etc. of coronary arteries.
In the prior art, intravascular ultrasound (intravenous ultrasound, IVUS) technology is generally used for intravascular-intimal edge detection, wherein intravascular ultrasound (intravenous ultrasound, IVUS) refers to a medical imaging technology that uses a special catheter with an ultrasound probe connected to its end, in combination with an invasive catheter technology.
However, since the IVUS image twists with the beating of the heart, the extraction of the lumen intima profile is not favored, resulting in a slower and less accurate extraction of the lumen intima profile of the vessel. Therefore, how to improve the accuracy of the vessel contour extraction is a technical problem that needs to be solved by those skilled in the art.
Based on the above, the embodiment of the application provides a method, a device, electronic equipment and a storage medium for extracting a blood vessel contour, which can effectively improve the speed and accuracy of extracting the blood vessel inner and outer diameter contours.
Referring to fig. 1, fig. 1 is a flowchart of a method for extracting a blood vessel contour according to an embodiment of the application. As shown in fig. 1, the extraction method provided by the embodiment of the application includes:
s101, acquiring a blood vessel section image set of a 3D blood vessel image to be identified.
Here, the blood vessel cross-sectional image set includes a plurality of blood vessel cross-sectional images to be identified, which are images taken according to a center line point on a blood vessel center line of the 3D blood vessel image to be identified.
Wherein, the characteristic of waiting to discern the blood vessel cross-section is: and (3) intercepting an original medical image with a certain size (including a blood vessel range) from a plane perpendicular to the tangential direction of the central line where the central line point is positioned.
The 3D blood vessel image to be identified can be a medical image such as CT, MRI and the like, and can be specifically a 3D blood vessel image at the position of coronary artery.
The vessel centerline is formed by a series of consecutive points (pixel space), and as an example, referring to fig. 2, fig. 2 is a schematic structural diagram of the vessel centerline according to the present application. The centerline points may be a plurality of points extracted from the centerline of the blood vessel at equal intervals or a plurality of points extracted from the centerline of the blood vessel according to a certain rule. The to-be-identified blood vessel cross section is a blood vessel cross section taken by central line points, each central line point can be used for taking a to-be-identified blood vessel cross section, and the number of images in the blood vessel cross section image set is determined by the number of selected central line points.
S102, inputting the blood vessel section image set into a pre-trained contour point identification model, and determining a first target polar coordinate point set of the inner diameter of a blood vessel and a second target polar coordinate point set of the outer diameter of the blood vessel in each blood vessel section image set to be identified.
The contour point recognition model can be a single-task recognition model or a multi-task recognition model.
When the contour point identification model is a single-task identification model, the output is a group of vectors formed by combining the inner diameter contour point and the outer diameter contour point, and then the first target polar coordinate point set of the inner diameter of the blood vessel and the second target polar coordinate point set of the outer diameter of the blood vessel can be obtained by dividing and decoding the combined vectors.
When the contour point identification model is a multitask identification model, two sets of vectors of the inner diameter contour point and the outer diameter contour point are output, and then a first target polar coordinate point set of the inner diameter of the blood vessel and a second target polar coordinate point set of the outer diameter of the blood vessel can be obtained by decoding the two sets of vectors.
The multi-task recognition model has the advantage of high accuracy compared with the single-task recognition model.
In one embodiment, the contour point identification model is constructed by:
s201, acquiring a sample blood vessel section image set of a plurality of sample 3D blood vessel images.
Here, each sample vessel cross-sectional view in the sample vessel cross-sectional image set corresponds to a first set of real polar points having an inside diameter of the vessel and a second set of real polar points having an outside diameter of the vessel.
The first real polar coordinate point set and the second real polar coordinate point set may be point sets formed by a preset number of coordinate points for discrete sampling.
The acquisition mode of the sample blood vessel cross-section image set is the same as that of the cross-section image set in step S101, and will not be described herein.
S202, sequentially inputting a sample blood vessel section image set of each sample 3D blood vessel image into an initial contour point identification neural network, and predicting a first predicted polar coordinate point set of the inner diameter of a blood vessel and a second predicted polar coordinate point set of the outer diameter of the blood vessel in each sample blood vessel section image.
Here, the initial contour point identification neural network may use a classical CNN network, or may use a more advanced specific network, which is not limited herein.
When the model is trained, independent training of the inner diameter contour points and the outer diameter contour points can be performed, and MultiClass, multiTask and the like can be adopted as training tasks; the method can also be used for combining and training the inner diameter contour points and the outer diameter contour points, respectively fixing the number of the inner diameter contour points and the outer diameter contour points, combining the inner diameter vector and the outer diameter vector into one 1 vector, and particularly can adopt a single-task vector regression method.
S203, determining a target loss function value based on the first real polar coordinate point set, the second real polar coordinate point set, the first predicted polar coordinate point set and the second predicted polar coordinate point set of each sample vessel cross-section.
Here, the objective loss function value is determined based on an error between the predicted value and the true value.
S204, based on the objective loss function value, performing iterative training on the initial contour point identification neural network, updating network parameters of the initial contour point identification neural network until the objective loss function value is converged, and stopping training to obtain the contour point identification model.
In one embodiment, for the polar coordinate point set described in step S201, the first real polar coordinate point set of the vessel inner diameter and the second real polar coordinate point set of the vessel outer diameter of each sample vessel cross-sectional view in the sample vessel cross-sectional image set may be specifically determined by:
s301, carrying out central line extraction processing on the sample 3D blood vessel image, and determining a blood vessel central line of the sample 3D blood vessel image.
Here, the center line extraction processing is performed on each sample 3D blood vessel image, and the blood vessel center line corresponding to each sample 3D blood vessel image is determined. Wherein the vessel centerline consists of a series of consecutive points (pixel space) comprising a plurality of centerline points on the vessel centerline, the centerline points being a plurality of target points previously selected from the vessel centerline.
S302, aiming at each central line point, cutting a plane with a preset size in the tangential direction of the central line of the vertical blood vessel comprising the central line point, and determining a section view of the sample blood vessel corresponding to the central line point.
Here, the preset size may be dynamically changed according to the thickness of the blood vessel, or may be a fixed value set in advance. Wherein the number of sample vessel cross-sectional views is the same as the number of selected centerline points.
S303, for each sample blood vessel section, based on the internal and external diameter contour labeling result on the sample blood vessel section, contour coordinate point extraction processing is performed, and a first real rectangular coordinate point set of the internal diameter of the blood vessel and a second real rectangular coordinate point set of the external diameter of the blood vessel in the sample blood vessel section are determined.
The marking result of the inner and outer diameter contours on the section view of the sample blood vessel is manual marking according to priori knowledge and/or gradient map after the section view of the sample blood vessel is determined, and the marking result of the contours can be two closed curves, namely an inner diameter and an outer diameter.
For example, referring to fig. 3, fig. 3 is a schematic structural diagram of a labeling result of an inner diameter profile and an outer diameter profile of a blood vessel according to the present application. Fig. 3 (a) is a schematic diagram of labeling results of inner and outer diameter profiles on a section view of a sample blood vessel by using an IVUS (intravascular ultrasound) technique based on priori knowledge, and fig. 3 (b) is a labeling result of inner and outer diameter profiles on a section view of a sample blood vessel based on a gradient map.
Thus, according to the labeling result, a first real rectangular coordinate point set of the inner diameter of the blood vessel and a second real rectangular coordinate point set of the outer diameter of the blood vessel in the section view of the blood vessel of the sample can be directly extracted. The set of rectangular coordinates points is also a set of cartesian coordinates points.
S304, carrying out coordinate conversion processing on the real polar coordinate point set in each sample blood vessel section view based on the central line point in the sample blood vessel section view, and determining the first real polar coordinate point set of the inner diameter of the blood vessel and the second real polar coordinate point set of the outer diameter of the blood vessel in the sample blood vessel section view.
Here, the coordinate conversion processing is performed by converting cartesian coordinate points into polar coordinate points.
The coordinate conversion process is performed because the following difficulties exist in performing contour prediction based on the cartesian coordinate system: 1. given learning information is disordered, and the background ratio is large: because of the different sizes of blood vessels, a large-sized cross-sectional view needs to be taken care of, and a lot of background interference is increased; the profile center information of the sectional views may not be uniform: the brightness of the angiography agent image is inconsistent under the condition of near segment and far segment and artifact; 2. the learning targets are more, and the learning targets comprise a contour point set consisting of a blood vessel contour center point and a blood vessel contour.
Converting rectangular coordinates (Cartesian coordinates) into polar coordinates simplifies the model training task, because the learned coordinate points (x, y) are converted into the learned radius r; thereby causing no excessive interference to the learned input information: background area reduction; reducing the learning objective: the (x, y) 2 features are learned, converted into a learning radius r, and can be reduced to a simple regression task.
For example, referring to fig. 4, fig. 4 is a schematic diagram illustrating a coordinate transformation principle provided by the present application. As shown in fig. 4, the left graph is a representation of contour points in a rectangular coordinate system, and the right graph is a representation of contour points in a polar coordinate system, where only one of the inside diameter or outside diameter is shown. Here, the spatial relationship of the cartesian coordinate system is converted into a relationship of the polar coordinates with respect to the rotation angle θ and the radius r; the rotation angle theta of the polar coordinates is sequentially unfolded, and the radius r is described as a continuous curve, so that only one parameter of the radius r is learned during model training, and the task is simplified.
In one embodiment, determining the objective loss function value based on the first set of real polar points, the second set of real polar points, the first set of predicted polar points, and the second set of predicted polar points for each sample vessel cross-sectional view described in step S203 includes:
S2031, determining a real distance set of the inner diameter of the blood vessel and the outer diameter of the blood vessel based on the first real polar coordinate point set and the second real polar coordinate point set.
The number of coordinate points in the first real polar coordinate point set and the second real polar coordinate point set may be the same or different. When the number of the inner and outer diameter distances is different, the same number of the inner and outer diameter distances is determined by taking the small number of the points as the main points.
S2032, determining a predicted distance set of the inner diameter and the outer diameter of the blood vessel based on the first predicted polar coordinate point set and the second predicted polar coordinate point set.
The determination method is the same as step S2031, and will not be described here again.
S2033, comparing the first predicted polar coordinate point set with the first real polar coordinate point set to determine a first loss function value.
S2034, comparing the second predicted polar coordinate point set with the second real polar coordinate point set to determine a second loss function value.
And S2035, comparing the predicted interval set with the real interval set to determine a third loss function value.
S2036 determining the objective loss function value based on the first loss function value, the second loss function value, and the third loss function value.
In one embodiment, the determining the objective loss function value in step S2036 based on the first loss function value, the second loss function value, and the third loss function value includes: and multiplying the first loss function value, the second loss function value and the third loss function value by the weight coefficients corresponding to the first loss function value, the second loss function value and the third loss function value respectively, and adding the three loss function values multiplied by the weight coefficients to determine the target loss function value.
The first, second, and third loss function values and the weight coefficients corresponding to the first, second, and third loss function values are preset. Here, the loss function value is calculated by increasing the relative distance between the inner diameter and the outer diameter, and the learning of the wall thickness of the blood vessel is enhanced.
S103, respectively carrying out coordinate conversion processing on a first target polar coordinate point set of the inner diameter of the blood vessel and a second target polar coordinate point set of the outer diameter of the blood vessel in each blood vessel section to be identified, and determining the first target rectangular coordinate point set of the inner diameter of the blood vessel and the second target rectangular coordinate point set of the outer diameter of the blood vessel in each blood vessel section to be identified.
Here, the value of the included angle between adjacent two points in the target polar coordinate point set is fixed. For example, assume that the first set of target polar coordinate points corresponding to each vessel cross-sectional view to be identified includes 60 points, and the included angle between two adjacent points is 6 °.
And S104, sequentially performing curve fitting treatment on a first target rectangular coordinate point set of the inner diameter of the blood vessel and a second target rectangular coordinate point set of the outer diameter of the blood vessel in each to-be-identified blood vessel section diagram, and determining a contour curve of the inner diameter of the blood vessel and a contour curve of the outer diameter of the blood vessel of the to-be-identified 3D blood vessel image.
In one embodiment, before performing curve fitting processing on the first target rectangular coordinate point set of the inner diameter of the blood vessel and the second target rectangular coordinate point set of the outer diameter of the blood vessel in sequence in each to-be-identified blood vessel cross-section, the extraction method further includes: and respectively carrying out interpolation processing on the first target rectangular coordinate point set and the second target rectangular coordinate point set, and increasing the number of coordinate points in the first target rectangular coordinate point set and the second target rectangular coordinate point set.
Here, interpolation processing is performed on the contour point set, so that the contour point is added to enrich the detail of the contour in order to be closer to the real form of the contour.
For example, referring to fig. 5, fig. 5 is a schematic structural diagram of an inner diameter contour line of a blood vessel according to the present application, and for better observation, only the inner diameter contour line is shown, and the outer diameter contour line is shown to be similar to the inner diameter contour line.
In addition, after the inner diameter contour line and the outer diameter contour line of the blood vessel are determined, the inner diameter contour curved surface and the outer diameter contour curved surface can be determined through the curved surface smoothing treatment.
Therefore, by converting rectangular coordinates into polar coordinates, background interference is reduced, model learning is simplified, and model learning speed is improved; the loss function is calculated through the distance between the inner diameter and the outer diameter, so that the study of the thickness of the blood vessel wall is enhanced, the model identification precision is improved, and the accuracy of extracting the inner diameter profile and the outer diameter profile of the blood vessel is further improved.
Referring to fig. 6 and 7, fig. 6 is a schematic structural diagram of a blood vessel profile extraction device according to an embodiment of the application, and fig. 7 is a schematic structural diagram of a blood vessel profile extraction device according to an embodiment of the application. As shown in fig. 6, the extraction apparatus 600 includes:
an acquisition module 610, configured to acquire a blood vessel cross-section image set of a 3D blood vessel image to be identified; the blood vessel section image set comprises a plurality of blood vessel section images to be identified, wherein the blood vessel section images to be identified are images intercepted according to a central line point on a blood vessel central line of the 3D blood vessel image to be identified;
a first determining module 620, configured to input the set of vessel cross-sectional images into a pre-trained contour point identification model, and determine a first target polar coordinate point set of an inner diameter of a vessel and a second target polar coordinate point set of an outer diameter of the vessel in each vessel cross-sectional image to be identified in the set of vessel cross-sectional images;
The coordinate conversion module 630 is configured to perform coordinate conversion processing on a first target polar coordinate point set of an inner diameter of a blood vessel and a second target polar coordinate point set of an outer diameter of the blood vessel in each of the to-be-identified blood vessel cross-sectional views, respectively, to determine a first target rectangular coordinate point set of the inner diameter of the blood vessel and a second target rectangular coordinate point set of the outer diameter of the blood vessel in each of the to-be-identified blood vessel cross-sectional views;
and the fitting processing module 640 is used for sequentially performing curve fitting processing on the first target rectangular coordinate point set of the inner diameter of the blood vessel and the second target rectangular coordinate point set of the outer diameter of the blood vessel in each to-be-identified blood vessel cross-section diagram, and determining a contour curve of the inner diameter of the blood vessel and a contour curve of the outer diameter of the blood vessel of the to-be-identified 3D blood vessel image.
Optionally, as shown in fig. 7, the extracting apparatus 600 further includes a model building module 650, where the model building module 650 is configured to:
acquiring a sample vessel cross-section image set of a plurality of sample 3D vessel images; wherein each sample vessel cross-sectional view in the sample vessel cross-sectional image set corresponds to a first set of real polar points having an inner vessel diameter and a second set of real polar points having an outer vessel diameter;
inputting a sample vessel section image set of each sample 3D vessel image into an initial contour point identification neural network in sequence, and predicting a first predicted polar coordinate point set of the inner diameter of a vessel and a second predicted polar coordinate point set of the outer diameter of the vessel in each sample vessel section image;
Determining a target loss function value based on the first real polar coordinate point set, the second real polar coordinate point set, the first predicted polar coordinate point set and the second predicted polar coordinate point set of each sample vessel cross-sectional view;
and carrying out iterative training on the initial contour point recognition neural network based on the objective loss function value, and updating network parameters of the initial contour point recognition neural network until the objective loss function value is converged, and stopping training to obtain the contour point recognition model.
Optionally, the extracting apparatus 600 further includes a second determining module 660, where the second determining module 660 is configured to:
performing central line extraction processing on the sample 3D blood vessel image, and determining a blood vessel central line of the sample 3D blood vessel image; wherein the vessel centerline comprises a plurality of centerline points thereon;
for each centerline point, intercepting a plane with a preset size in the tangential direction of the centerline of the perpendicular blood vessel comprising the centerline point, and determining a sample blood vessel section diagram corresponding to the centerline point;
for each sample blood vessel section, based on the internal and external diameter contour labeling result on the sample blood vessel section, performing contour coordinate point extraction processing to determine a first real rectangular coordinate point set of the internal diameter of the blood vessel and a second real rectangular coordinate point set of the external diameter of the blood vessel in the sample blood vessel section;
And carrying out coordinate conversion processing on the basis of the central line points in the section of each sample blood vessel according to the real polar coordinate point set in the section of each sample blood vessel, and determining a first real polar coordinate point set of the inner diameter of the blood vessel and a second real polar coordinate point set of the outer diameter of the blood vessel in the section of each sample blood vessel.
Optionally, the model building module 650, when configured to determine the objective loss function value based on the first set of real polar points, the second set of real polar points, the first set of predicted polar points, and the second set of predicted polar points for each sample vessel cross-sectional view, the model building module 650 is configured to:
determining a real interval set of the inner diameter of the blood vessel and the outer diameter of the blood vessel based on the first real polar coordinate point set and the second real polar coordinate point set;
determining a predicted distance set of the inner diameter of the blood vessel and the outer diameter of the blood vessel based on the first predicted polar coordinate point set and the second predicted polar coordinate point set;
comparing the first predicted polar coordinate point set with the first real polar coordinate point set to determine a first loss function value;
comparing the second predicted polar coordinate point set with the second real polar coordinate point set to determine a second loss function value;
Comparing the predicted distance set with the real distance set to determine a third loss function value;
the objective loss function value is determined based on the first loss function value, the second loss function value, and the third loss function value.
Optionally, the model building module 650, when configured to determine the objective loss function value based on the first loss function value, the second loss function value, and the third loss function value, is configured to:
and multiplying the first loss function value, the second loss function value and the third loss function value by the weight coefficients corresponding to the first loss function value, the second loss function value and the third loss function value respectively, and adding the three loss function values multiplied by the weight coefficients to determine the target loss function value.
Optionally, the extracting apparatus 600 further includes an interpolation module 670, where the interpolation module 670 is configured to: before curve fitting processing is sequentially carried out on a first target rectangular coordinate point set of the inner diameter of a blood vessel and a second target rectangular coordinate point set of the outer diameter of the blood vessel in each blood vessel section diagram to be identified, interpolation processing is respectively carried out on the first target rectangular coordinate point set and the second target rectangular coordinate point set, and the number of coordinate points in the first target rectangular coordinate point set and the second target rectangular coordinate point set is increased.
Optionally, the contour point recognition model is a single-task recognition model or a multi-task recognition model.
Referring to fig. 8, fig. 8 is a schematic structural diagram of an electronic device according to an embodiment of the application. As shown in fig. 8, the electronic device 800 includes a processor 810, a memory 820, and a bus 830.
The memory 820 stores machine-readable instructions executable by the processor 810, and when the electronic device 800 is running, the processor 810 and the memory 820 communicate through the bus 830, and when the machine-readable instructions are executed by the processor 810, the steps in the method embodiments shown in fig. 1 to 5 can be executed, and the specific implementation can be referred to the method embodiments, which are not repeated herein.
The embodiment of the present application further provides a computer readable storage medium, where a computer program is stored, where the computer program may execute the steps in the method embodiments shown in the foregoing fig. 1 to 5 when the computer program is executed by a processor, and a specific implementation manner may refer to the method embodiments and is not repeated herein.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, and are not repeated herein.
In the several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. The above-described apparatus embodiments are merely illustrative, for example, the division of the units is merely a logical function division, and there may be other manners of division in actual implementation, and for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some communication interface, device or unit indirect coupling or communication connection, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer readable storage medium executable by a processor. Based on this understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
Finally, it should be noted that: the above examples are only specific embodiments of the present application, and are not intended to limit the scope of the present application, but it should be understood by those skilled in the art that the present application is not limited thereto, and that the present application is described in detail with reference to the foregoing examples: any person skilled in the art may modify or easily conceive of the technical solution described in the foregoing embodiments, or perform equivalent substitution of some of the technical features, while remaining within the technical scope of the present disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application, and are intended to be included in the scope of the present application. Therefore, the protection scope of the application is subject to the protection scope of the claims.
Claims (10)
1. A method of extracting a blood vessel contour, the method comprising:
acquiring a blood vessel section image set of a 3D blood vessel image to be identified; the blood vessel section image set comprises a plurality of blood vessel section images to be identified, wherein the blood vessel section images to be identified are images intercepted according to a central line point on a blood vessel central line of the 3D blood vessel image to be identified;
inputting the blood vessel section image set into a pre-trained contour point identification model, and determining a first target polar coordinate point set of the inner diameter of a blood vessel and a second target polar coordinate point set of the outer diameter of the blood vessel in each blood vessel section image set to be identified;
respectively carrying out coordinate conversion treatment on a first target polar coordinate point set of the inner diameter of the blood vessel and a second target polar coordinate point set of the outer diameter of the blood vessel in each blood vessel section to be identified, and determining a first target rectangular coordinate point set of the inner diameter of the blood vessel and a second target rectangular coordinate point set of the outer diameter of the blood vessel in each blood vessel section to be identified;
and sequentially performing curve fitting treatment on a first target rectangular coordinate point set of the inner diameter of the blood vessel and a second target rectangular coordinate point set of the outer diameter of the blood vessel in each to-be-identified blood vessel section view, and determining a contour curve of the inner diameter of the blood vessel and a contour curve of the outer diameter of the blood vessel of the to-be-identified 3D blood vessel image.
2. The extraction method according to claim 1, wherein the contour point recognition model is constructed by:
acquiring a sample vessel cross-section image set of a plurality of sample 3D vessel images; wherein each sample vessel cross-sectional view in the sample vessel cross-sectional image set corresponds to a first set of real polar points having an inner vessel diameter and a second set of real polar points having an outer vessel diameter;
inputting a sample vessel section image set of each sample 3D vessel image into an initial contour point identification neural network in sequence, and predicting a first predicted polar coordinate point set of the inner diameter of a vessel and a second predicted polar coordinate point set of the outer diameter of the vessel in each sample vessel section image;
determining a target loss function value based on the first real polar coordinate point set, the second real polar coordinate point set, the first predicted polar coordinate point set and the second predicted polar coordinate point set of each sample vessel cross-sectional view;
and carrying out iterative training on the initial contour point recognition neural network based on the objective loss function value, and updating network parameters of the initial contour point recognition neural network until the objective loss function value is converged, and stopping training to obtain the contour point recognition model.
3. The extraction method according to claim 2, characterized in that the first set of true polar points of the vessel inner diameter and the second set of true polar points of the vessel outer diameter of each sample vessel cross-sectional view in the set of sample vessel cross-sectional images are determined by:
performing central line extraction processing on the sample 3D blood vessel image, and determining a blood vessel central line of the sample 3D blood vessel image; wherein the vessel centerline comprises a plurality of centerline points thereon;
for each centerline point, intercepting a plane with a preset size in the tangential direction of the centerline of the perpendicular blood vessel comprising the centerline point, and determining a sample blood vessel section diagram corresponding to the centerline point;
for each sample blood vessel section, based on the internal and external diameter contour labeling result on the sample blood vessel section, performing contour coordinate point extraction processing to determine a first real rectangular coordinate point set of the internal diameter of the blood vessel and a second real rectangular coordinate point set of the external diameter of the blood vessel in the sample blood vessel section;
and carrying out coordinate conversion processing on the basis of the central line points in the section of each sample blood vessel according to the real polar coordinate point set in the section of each sample blood vessel, and determining a first real polar coordinate point set of the inner diameter of the blood vessel and a second real polar coordinate point set of the outer diameter of the blood vessel in the section of each sample blood vessel.
4. The method of extraction of claim 1, wherein determining the objective loss function value based on the first set of true polar points, the second set of true polar points, the first set of predicted polar points, and the second set of predicted polar points for each sample vessel cross-sectional view comprises:
determining a real interval set of the inner diameter of the blood vessel and the outer diameter of the blood vessel based on the first real polar coordinate point set and the second real polar coordinate point set;
determining a predicted distance set of the inner diameter of the blood vessel and the outer diameter of the blood vessel based on the first predicted polar coordinate point set and the second predicted polar coordinate point set;
comparing the first predicted polar coordinate point set with the first real polar coordinate point set to determine a first loss function value;
comparing the second predicted polar coordinate point set with the second real polar coordinate point set to determine a second loss function value;
comparing the predicted distance set with the real distance set to determine a third loss function value;
the objective loss function value is determined based on the first loss function value, the second loss function value, and the third loss function value.
5. The extraction method according to claim 4, wherein the determining the objective loss function value based on the first loss function value, the second loss function value, and the third loss function value includes:
And multiplying the first loss function value, the second loss function value and the third loss function value by the weight coefficients corresponding to the first loss function value, the second loss function value and the third loss function value respectively, and adding the three loss function values multiplied by the weight coefficients to determine the target loss function value.
6. The extraction method according to claim 1, wherein before sequentially performing curve fitting processing on the first set of target rectangular coordinates of the inside diameter of the blood vessel and the second set of target rectangular coordinates of the outside diameter of the blood vessel in each of the blood vessel sectional views to be identified, the extraction method further comprises:
and respectively carrying out interpolation processing on the first target rectangular coordinate point set and the second target rectangular coordinate point set, and increasing the number of coordinate points in the first target rectangular coordinate point set and the second target rectangular coordinate point set.
7. The extraction method according to claim 1, wherein the contour point recognition model is a single-task recognition model or a multi-task recognition model.
8. An extraction device of a blood vessel contour, characterized in that the extraction device comprises:
the acquisition module is used for acquiring a blood vessel section image set of the 3D blood vessel image to be identified; the blood vessel section image set comprises a plurality of blood vessel section images to be identified, wherein the blood vessel section images to be identified are images intercepted according to a central line point on a blood vessel central line of the 3D blood vessel image to be identified;
The first determining module is used for inputting the blood vessel section image set into a pre-trained contour point identification model and determining a first target polar coordinate point set of the inner diameter of a blood vessel and a second target polar coordinate point set of the outer diameter of the blood vessel in each blood vessel section image set to be identified;
the coordinate conversion module is used for carrying out coordinate conversion treatment on a first target polar coordinate point set of the inner diameter of the blood vessel and a second target polar coordinate point set of the outer diameter of the blood vessel in each blood vessel section diagram to be identified respectively, and determining a first target rectangular coordinate point set of the inner diameter of the blood vessel and a second target rectangular coordinate point set of the outer diameter of the blood vessel in each blood vessel section diagram to be identified;
and the fitting processing module is used for sequentially carrying out curve fitting processing on a first target rectangular coordinate point set of the inner diameter of the blood vessel and a second target rectangular coordinate point set of the outer diameter of the blood vessel in each to-be-identified blood vessel section diagram, and determining a contour curve of the inner diameter of the blood vessel and a contour curve of the outer diameter of the blood vessel of the to-be-identified 3D blood vessel image.
9. An electronic device, comprising: a processor, a memory and a bus, said memory storing machine readable instructions executable by said processor, said processor and said memory communicating via said bus when the electronic device is running, said machine readable instructions when executed by said processor performing the steps of the extraction method according to any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a computer program which, when executed by a processor, performs the steps of the extraction method according to any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310837553.1A CN116862874A (en) | 2023-07-10 | 2023-07-10 | Blood vessel contour extraction method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310837553.1A CN116862874A (en) | 2023-07-10 | 2023-07-10 | Blood vessel contour extraction method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116862874A true CN116862874A (en) | 2023-10-10 |
Family
ID=88231768
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310837553.1A Pending CN116862874A (en) | 2023-07-10 | 2023-07-10 | Blood vessel contour extraction method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116862874A (en) |
-
2023
- 2023-07-10 CN CN202310837553.1A patent/CN116862874A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20210272030A1 (en) | Systems and methods for using geometry sensitivity information for guiding workflow | |
| Benes et al. | Automatically designed machine vision system for the localization of CCA transverse section in ultrasound images | |
| KB et al. | Convolutional neural network for segmentation and measurement of intima media thickness | |
| CN111640124B (en) | Blood vessel extraction method, device, equipment and storage medium | |
| CN113160189A (en) | Blood vessel center line extraction method, device, equipment and storage medium | |
| Han et al. | Automatic coronary artery segmentation using active search for branches and seemingly disconnected vessel segments from coronary CT angiography | |
| WO2016159379A1 (en) | Apparatus and method for constructing blood vessel configuration and computer software program | |
| CN112150454B (en) | Aortic dissection assessment method, device, equipment and storage medium | |
| CN114004793A (en) | Blood flow characteristic acquisition method and device, electronic equipment and storage medium | |
| US10354349B2 (en) | Systems and methods for using geometry sensitivity information for guiding workflow | |
| CN117036530B (en) | Cross-modal data-based coronary artery fractional flow reserve prediction method and device | |
| CN112651984A (en) | Blood vessel lumen intimal contour extraction method and device, ultrasonic equipment and storage medium | |
| CN116862874A (en) | Blood vessel contour extraction method and device, electronic equipment and storage medium | |
| CN116138877A (en) | Target positioning method, target positioning device, electronic equipment and storage medium | |
| CN114170258A (en) | Image segmentation method and device, electronic equipment and storage medium | |
| CN112001893B (en) | Calculation method, device and equipment of vascular parameters and storage medium | |
| CN115482261A (en) | Blood vessel registration method, device, electronic equipment and storage medium | |
| JP2023131287A (en) | Information processing apparatus and learning method | |
| Chalopin et al. | Evaluation of a semi-automatic segmentation algorithm in 3D intraoperative ultrasound brain angiography | |
| CN111588409B (en) | Super-resolution reconstruction method and device for three-dimensional ultrasonic contrast image | |
| Wang et al. | A new deformable contour method | |
| Zhang et al. | Vascular centerline extraction of CTA images based on minimal path and Bayesian tracking | |
| CN116630634B (en) | Image processing method, device, equipment and storage medium | |
| Sabrowsky-Hirsch et al. | A content-driven architecture for medical image segmentation | |
| CN116883358A (en) | Vascular image processing method and device, electronic equipment and storage medium |
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 |