CN111325759B

CN111325759B - Vessel segmentation method, apparatus, computer device, and readable storage medium

Info

Publication number: CN111325759B
Application number: CN202010177262.0A
Authority: CN
Inventors: 宋燕丽; 杨帆; 吴迪嘉
Original assignee: Lianying Intelligent Medical Technology Beijing Co ltd; Shanghai United Imaging Intelligent Healthcare Co Ltd
Current assignee: Lianying Intelligent Medical Technology Beijing Co ltd; Shanghai United Imaging Intelligent Healthcare Co Ltd
Priority date: 2020-03-13
Filing date: 2020-03-13
Publication date: 2024-04-16
Anticipated expiration: 2040-03-13
Also published as: CN111325759A

Abstract

The application relates to a blood vessel segmentation method, a blood vessel segmentation device, a computer device and a storage medium. The method comprises the following steps: acquiring an image to be detected; inputting the image to be detected into a tissue organ segmentation network to obtain a tissue organ image; acquiring a distance field of a blood vessel and a tissue organ corresponding to the blood vessel according to the tissue organ image; determining an annular band containing the blood vessel from the distance field and tissue organ images; inputting the annular band into the blood vessel segmentation network to obtain an initial blood vessel image; extracting a first image block in a preset range around a blood vessel center line endpoint according to the initial blood vessel image; and inputting the first image block into a vessel tracking network for end point tracking to obtain a final vessel image. The accuracy and the stability of the acquired blood vessel image can be improved by the method, and the image is corrected by the neural network, so that the time and the energy of technicians are greatly saved, and the diagnosis cost is further saved.

Description

Vessel segmentation method, apparatus, computer device, and readable storage medium

Technical Field

The present application relates to the field of medical imaging, and in particular, to a blood vessel segmentation method, apparatus, computer device, and readable storage medium.

Background

Cardiovascular diseases are diseases with high morbidity and mortality, wherein common diseases such as coronary artery stenosis and plaque seriously jeopardize human health. The cardiovascular disease has the characteristics of urgent onset, strong concealment and the like, so that the diagnosis of the heart disease is realized with great clinical significance. Because coronary vessels are of an elongated tubular structure and have large variability, clinical diagnosis of coronary vessels generally depends on the segmentation result of the coronary vessels, namely, coronary arteries are automatically segmented, and then the coronary vessels are diagnosed by adopting CPR reconstruction (curved surface reconstruction), coronary probe reconstruction, VR and other visual technologies on the basis of segmentation, so that basis is provided for early prevention and diagnosis of cardiovascular diseases by doctors.

The coronary segmentation technique commonly used at present is generally based on a conventional image processing method, such as linear structure detection, deformation model or a conventional machine learning method. However, these methods generally have poor accuracy and stability of segmentation, and require a technician to spend a lot of time and effort to manually correct and edit, which greatly affects the diagnosis of coronary artery disease by the doctor.

Disclosure of Invention

The embodiment of the application provides a blood vessel segmentation method, a blood vessel segmentation device, computer equipment and a readable storage medium, which are used for at least solving the problem of poor accuracy and stability of the related technology.

A method of vessel segmentation, comprising: acquiring an image to be detected; inputting the image to be detected into a tissue organ segmentation network to obtain a tissue organ image; acquiring a distance field of a blood vessel and a tissue organ corresponding to the blood vessel according to the tissue organ image; determining an annular band containing the blood vessel from the distance field and tissue organ images; inputting the annular band into the blood vessel segmentation network to obtain an initial blood vessel image; extracting a first image block in a preset range around a blood vessel center line endpoint according to the initial blood vessel image; and inputting the first image block into a vessel tracking network for end point tracking to obtain a final vessel image.

In one embodiment, inputting the annular band into the vessel segmentation network, obtaining an initial vessel image includes: dividing the annular belt into a plurality of second image blocks according to a preset size; respectively inputting a plurality of second image blocks into the blood vessel segmentation network to obtain a plurality of blood vessel segmentation images; and splicing the plurality of blood vessel segmentation images to obtain an initial blood vessel image.

In one embodiment, the stitching the plurality of vessel segmentation images to obtain an initial vessel image includes: the second image block includes a position code; acquiring the position code of each blood vessel segmentation image in an annular band according to the position code of each second image block; and filling each blood vessel segmentation image to a position corresponding to the position code according to the position code of the blood vessel segmentation image, so as to obtain an initial blood vessel image.

In one embodiment, extracting a first image block within a preset range around a vessel centerline endpoint from the initial vessel image includes: extracting a blood vessel center line according to the initial blood vessel image; searching the end point of the blood vessel center line according to the blood vessel center line; and taking the image in a preset range around the center line end point of the blood vessel as a first image block.

In one embodiment, inputting the first image block into a vessel tracking network for end point tracking, and obtaining a final vessel image includes: inputting the first image block into a vessel tracking network to perform end point tracking to obtain an end point tracking image; extracting a connected domain according to the endpoint tracking image; and splicing the connected domain to the initial blood vessel image to obtain a final blood vessel image.

In one embodiment, the stitching the connected domain to the initial blood vessel image to obtain a final blood vessel image includes: splicing the connected domain to an initial blood vessel image, and counting the splicing times; if the splicing times are greater than or equal to the preset times, obtaining a final blood vessel image; and if the splicing times are smaller than the preset times, extracting the first image blocks in the preset range around the blood vessel center line end point again to carry out end point tracking until the splicing times are larger than or equal to the preset times, and obtaining a final blood vessel image.

In one embodiment, the stitching the connected domain to the initial blood vessel image to obtain a final blood vessel image includes: splicing the connected domain to the initial blood vessel image, and acquiring a first length of a blood vessel center line in an end point tracking image and a second length of the blood vessel center line in a first image block corresponding to the end point tracking image; the first length and the second length are subjected to difference to obtain an increasing distance; if the growing distance is smaller than the preset distance, obtaining a final blood vessel image; and if the growing distance is greater than or equal to the preset distance, extracting the first image blocks in the preset range around the center line end point of the blood vessel again to track the blood vessel until the growing distance is smaller than the preset distance, and obtaining a final blood vessel image.

A vessel segmentation device, comprising: the acquisition module is used for acquiring the image to be detected; the tissue organ segmentation module is used for inputting the image to be detected into a tissue organ segmentation network to obtain a tissue organ image; the distance field calculation module is used for acquiring a distance field of a blood vessel and a tissue organ corresponding to the blood vessel according to the tissue organ image; an annular band calculation module for determining an annular band containing the blood vessel based on the distance field and tissue organ images; the blood vessel segmentation module is used for inputting the annular belt into the blood vessel segmentation network to obtain an initial blood vessel image; the endpoint extraction module is used for extracting a first image block in a preset range around the endpoint of the blood vessel center line according to the initial blood vessel image; and the end point tracking module is used for inputting the first image block into a blood vessel tracking network for end point tracking to obtain a final blood vessel image.

A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing any of the vessel segmentation methods described above when executing the computer program.

A computer readable storage medium having stored thereon a computer program which when executed by a processor implements any of the above described vessel segmentation methods.

Compared with the related art, the blood vessel segmentation method provided by the embodiment of the application comprises the steps of obtaining an image to be detected, inputting the image to be detected into a tissue organ segmentation network to obtain a tissue organ image, obtaining a distance field of a blood vessel and a tissue organ corresponding to the blood vessel according to the tissue organ image, determining an annular band containing the blood vessel according to the distance field and the tissue organ image, inputting the annular band into the blood vessel segmentation network to obtain an initial blood vessel image, extracting a first image block in a preset range around a center line end point of the blood vessel according to the initial blood vessel image, and finally inputting the first image block into a blood vessel tracking network to carry out end point tracking to obtain a final blood vessel image. The accuracy and the stability of the acquired blood vessel image can be improved by the method, and the image is corrected by the neural network, so that the time and the energy of technicians are greatly saved, and the diagnosis cost is further saved.

The details of one or more embodiments of the application are set forth in the accompanying drawings and the description below to provide a more thorough understanding of the other features, objects, and advantages of the application.

Drawings

FIG. 1 is a flow chart of a method of vessel segmentation in one embodiment;

FIG. 2 is a schematic diagram of a network architecture in one embodiment;

FIG. 3 is a schematic diagram of a network structure according to another embodiment;

FIG. 4 is a flow chart of a method of acquiring an initial vessel image in one embodiment;

FIG. 5 is a flow chart of a method of acquiring a final blood vessel image in one embodiment;

FIG. 6 is a flow chart of a method for acquiring a final blood vessel image according to another embodiment;

FIG. 7 is a process flow of coronary segmentation in one embodiment;

FIG. 8 is a heart split image according to one embodiment;

FIG. 9 is a schematic view of an endless belt in one embodiment;

FIG. 10 (a) is an initial coronary segmentation effect graph, FIG. 10 (b) is a coronary tracking effect graph, and FIG. 10 (c) is a coronary segmentation golden standard label;

FIG. 11 (a) is an input image block of a coronary segmented network; FIG. 11 (b) is a graph of initial coronary segmentation results output by the coronary segmentation network; FIG. 11 (c) is a coronary centerline extracted based on FIG. 11 (b); FIG. 11 (d) is an input coronary endpoint image block of the coronary tracking network; FIG. 11 (e) is a coronary tracking image output by the coronary tracking network; FIG. 11 (f) is an image obtained by stitching the coronary tracking image to the initial coronary segmentation result;

FIG. 12 is a graph showing a comparison of coronary segmentation effects in one embodiment;

FIG. 13 is a graph showing comparison of coronary segmentation results with a golden standard in one embodiment;

FIG. 14 is a block diagram of a vascular segmentation device in one embodiment;

fig. 15 is an internal structural view of a computer device in one embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described and illustrated below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application. All other embodiments, which can be made by one of ordinary skill in the art without undue burden on the person of ordinary skill in the art based on the embodiments provided herein, are intended to be within the scope of the present application.

It is apparent that the drawings in the following description are only some examples or embodiments of the present application, and it is possible for those of ordinary skill in the art to apply the present application to other similar situations according to these drawings without inventive effort. Moreover, it should be appreciated that while such a development effort might be complex and lengthy, it would nevertheless be a routine undertaking of design, fabrication, or manufacture for those of ordinary skill having the benefit of this disclosure, and thus should not be construed as having the benefit of this disclosure.

Reference in the specification to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is to be expressly and implicitly understood by those of ordinary skill in the art that the embodiments described herein can be combined with other embodiments without conflict.

Unless defined otherwise, technical or scientific terms used herein should be given the ordinary meaning as understood by one of ordinary skill in the art to which this application belongs. Reference to "a," "an," "the," and similar terms herein do not denote a limitation of quantity, but rather denote the singular or plural. The terms "comprising," "including," "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, article, or apparatus that comprises a list of steps or modules (elements) is not limited to only those steps or elements but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus. The terms "connected," "coupled," and the like in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The term "plurality" as used herein refers to two or more. "and/or" describes an association relationship of an association object, meaning that there may be three relationships, e.g., "a and/or B" may mean: a exists alone, A and B exist together, and B exists alone. The character "/" generally indicates that the context-dependent object is an "or" relationship. The terms "first," "second," "third," and the like, as used herein, are merely distinguishing between similar objects and not representing a particular ordering of objects.

In order to acquire an image to be measured of a scanning object, the scanning object needs to be scanned by using a medical imaging device, wherein the scanning object can be a whole organ of a patient, or can be an organ, a tissue or a cell collection which needs to be detected by the patient in a key way, and the like. The medical imaging device scans the scanned object to obtain scanning data, and a medical image sequence is generated according to the scanning data. Wherein the medical image sequence is an image of each cross section of the scan object in the scan direction. And finally generating a three-dimensional image of the internal structure of the scanning object according to the image sequence. Wherein the medical imaging device may be: x-ray imaging apparatus, CT (normal CT, spiral CT), positive sub-scan (PET), magnetic resonance imaging (MR), infrared scanning apparatus, and a combination scanning apparatus of a plurality of scanning apparatuses.

The following is exemplified by a medical imaging device as CT: a computed tomography apparatus (CT) generally comprises a gantry, a scan table and a console for operation by a physician. One side of the frame is provided with a bulb, and one side opposite to the bulb is provided with a detector. The console is computer equipment for controlling the bulb tube and the detector to scan, and the computer equipment is also used for receiving the data acquired by the detector, processing and reconstructing the data, and finally forming a CT image. When the CT is used for scanning, a patient lies on the scanning bed, the scanning bed sends the patient into the aperture of the frame, the bulb tube arranged on the frame emits X rays, the X rays penetrate through the patient to be received by the detector to form data, the data are transmitted to the computer equipment, and the computer equipment performs preliminary processing and image reconstruction on the data to obtain CT images.

The embodiment also provides a blood vessel segmentation method. FIG. 1 is a schematic flow chart of a method for segmenting a blood vessel in one embodiment, as shown in FIG. 1, the flow chart includes the following steps:

step S101, acquiring an image to be measured.

Specifically, the medical imaging device scans a human body to be scanned first to obtain an image to be measured. The acquired image to be measured can be directly scanned by medical imaging equipment to be scanned on a human body and directly reconstructed to obtain the image to be measured; the obtaining of the image to be detected can also be that the medical imaging equipment scans the human body to be scanned, stores the scanning data, obtains the stored scanning data and reconstructs the stored scanning data to obtain the image to be detected; the obtaining of the image to be detected can also be that the medical imaging equipment scans the human body to be scanned to reconstruct to obtain the image to be detected, and the image to be detected is stored to obtain the stored image to be detected. The image to be measured is an image obtained after image reconstruction of the scanning data.

Step S102, inputting the image to be detected into a tissue organ segmentation network to obtain a tissue organ image.

Specifically, the tissue organ segmentation network adopts two neural networks from low resolution to high resolution through a deep learning method. The obtained image to be measured is input into a tissue organ segmentation network to obtain tissue organ images, for example, the tissue organs such as stomach, lung, muscle and the like can be obviously extracted from the images.

Step S103, acquiring a distance field of a blood vessel and a tissue organ corresponding to the blood vessel according to the tissue organ image.

Specifically, a blood vessel exists in or near a tissue organ, firstly, the blood vessel is found in the tissue organ image, if the blood vessel is outside the tissue organ, namely near the tissue organ, the minimum distance from all pixel points of the tissue organ wall to the current blood vessel point is calculated, and all blood vessel points are traversed to obtain the extraluminal distance; if the blood vessel is in the tissue organ, namely in the tissue organ, calculating the minimum distance from all pixel points of the tissue organ wall to the current blood vessel point, and traversing all blood vessel points to obtain the intracavity distance. Wherein the distance field comprises an extraluminal distance of the blood vessel outside the tissue organ and an intracavitary distance of the blood vessel inside the tissue organ. The distance field is used to represent the distance of a blood vessel to the wall of a tissue organ.

Step S104, determining an annular band containing the blood vessel according to the distance field and the tissue organ image.

Specifically, according to the intra-cavity distance and the extra-cavity distance of the distance field, determining the annular zone where the blood vessel is located in the tissue organ image, determining the farthest distance from the tissue organ wall when the blood vessel is located outside the tissue organ through the extra-cavity distance, and determining the farthest distance from the tissue organ wall when the blood vessel is located in the tissue organ through the intra-cavity distance, thereby determining the annular zone where the blood vessel is located, namely the annular zone. The probability of occurrence of blood vessels in the annular band is very large, while the probability of occurrence of blood vessels out of the annular band is very small.

Step S105, inputting the annular band into the blood vessel segmentation network to obtain an initial blood vessel image.

Specifically, the vascular segmentation network is a neural network obtained through a deep learning method, the vascular segmentation network comprises 4 times of up-sampling and 4 times of down-sampling, and comprises a characteristic extraction process of 5 scales, wherein the number of residual modules is 1-5. The input of the blood vessel segmentation network is a gray image in the annular band, namely, a tissue organ image in the annular band, and the output is an initial blood vessel image. The initial blood vessel image is an image showing the blood vessel.

Step S106, extracting a first image block in a preset range around the center line end point of the blood vessel according to the initial blood vessel image.

Specifically, extracting a blood vessel center line according to the initial blood vessel image; searching the end point of the blood vessel center line according to the blood vessel center line; and taking the image in a preset range around the center line end point of the blood vessel as a first image block. More specifically, from the initial blood vessel image, a blood vessel center line in the initial blood vessel image is first extracted, and a plurality of end points of the blood vessel center line are searched. For each endpoint, an image within a preset range around the endpoint is first extracted as a first image block. Wherein the preset range may be an image within 5mm around the end point as the first image block. The preset range can be set according to the scene requirement in actual use, and the embodiment is not particularly limited. The central line of the blood vessel is a line formed by connecting blood vessel midpoints in the initial blood vessel image.

And step S107, inputting the first image block into a vessel tracking network for end point tracking to obtain a final vessel image.

Specifically, the vessel tracking network is a neural network obtained through a deep learning method, and includes 2 upsampling and 2 downsampling, and focuses on feature extraction of a local image. Inputting a first image block comprising an endpoint into a vessel tracking network to perform endpoint tracking to obtain an endpoint tracking image, extracting a connected domain from the endpoint tracking image, splicing the connected domain to an initial vessel image, splicing newly tracked pixel points to the initial vessel image, and obtaining a final vessel image. On the basis of carrying out primary endpoint tracking, multiple times of endpoint tracking can be carried out, and a final image is obtained through splicing.

According to the blood vessel segmentation method provided by the embodiment, firstly, a tissue organ image is obtained through a tissue organ segmentation network, then a distance field is determined through the tissue organ image, an annular band where blood vessels are located is determined according to the distance field and the tissue organ image, an initial blood vessel image in the annular band is determined through the blood vessel segmentation network, and finally, the end points of all blood vessels in the initial blood vessel image are tracked through a blood vessel tracking network, so that a final blood vessel image is obtained. The accuracy and the stability of the acquired blood vessel image can be improved by the method, and the image is corrected by the neural network, so that the time and the energy of technicians are greatly saved, and the diagnosis cost is further saved.

In one embodiment, the tissue organ segmentation network, the vessel segmentation network, and the vessel tracking network may be constructed based on the following: convolutional Neural Network (CNN), generating an antagonism network (GAN), or the like, or a combination thereof. Examples of Convolutional Neural Networks (CNNs) may include SRCNN (Super-Resolution Convolutional Neural Network, super-resolution convolutional neural networks), dnCNN (Denoising Convolutional Neural Network, denoising convolutional neural networks), U-net, V-net, and FCN (Fully Convolutional Network, full convolutional neural networks). In some embodiments, the neural network model may include multiple layers, such as an input layer, multiple hidden layers, and an output layer. The plurality of hidden layers may include one or more convolution layers, one or more batch normalization layers, one or more activation layers, a full connection layer, a cost function layer, and the like. Each of the plurality of layers may include a plurality of nodes.

In one embodiment, the tissue organ segmentation network and the blood vessel segmentation network may be network structures as shown in fig. 2. Specifically, the network structure comprises 4 downsampling and 4 upsampling, and comprises a feature extraction process of 5 scales, wherein the number of residual modules is selectable from 1 to 5. The neural network may cut image blocks of 64-256 pixels in size, preferably 128 x 128, during training. The neural network cuts images may have a resolution of 0.3-0.8mm, preferably 0.6mm. When the neural network performs normalization processing on the image, the normalization processing is performed in a mean value and standard deviation mode, wherein the preferred mean value is 50, and the standard deviation is 400. The neural network adopts minimization of the dice-loss coefficient as an optimization target. The neural network adopts an Adam parameter optimization method.

In one embodiment, the vessel tracking network may be a network structure as shown in FIG. 3. Specifically, the network architecture includes 2 upsamples and 2 downsamples, focusing on feature extraction of the partial image. The neural network may cut image blocks of 8-32 pixels in size, preferably 16 x 16, during training. The neural network may cut an image with a resolution of 0.3-0.8mm, preferably 0.6mm. When the neural network performs normalization processing on the image, the normalization processing is performed in a mean value and standard deviation mode, wherein the preferred mean value is 50, and the standard deviation is 400. The neural network adopts minimization of the dice-loss coefficient as an optimization target. The neural network adopts an Adam parameter optimization method. The neural network is sampled in a positive sample mode on a blood vessel and in a negative sample mode in an area 5mm outside the blood vessel. Through the network structure shown in fig. 3, the conventional network structure sets up multiple samplings, and the embodiment reduces the calculation amount of sampling by only setting up sampling for 2 times and down sampling for 2 times, can satisfy the feature extraction of the local image, and can also ensure the accuracy of the local feature extraction.

In one embodiment, a method for acquiring an initial blood vessel image is provided, and fig. 4 is a schematic flow chart of the method for acquiring an initial blood vessel image in one embodiment, as shown in fig. 4, and the flow chart includes the following steps:

step S401: the endless belt is divided into a plurality of second image blocks according to a preset size.

Specifically, a point is selected on the annular belt, and a second image block is sequentially and continuously intercepted by taking the point as a starting point and a preset step length and a preset dicing size. Wherein the step size represents how far apart a cube is sampled; the dicing size refers to the size of the cut image block. The step size needs to be less than or equal to the size of the cut-out so that the image in the entire annular band is cut out, preventing that a partial area is not cut out. The smaller the step size, the denser the sampling density of the image blocks, the larger the overlap area between the cut individual image blocks, and the slower the cutting speed of the entire endless belt. Preferably, the step size is consistent with the dicing size, i.e. there is no overlap between adjacent two image blocks. Such a cutting method can maximize the cutting speed when all the images in the endless belt are cut. The preset step length is 10mm at minimum, and the preset step length is 10mm, and the dicing size is 10mm for example, so as to exemplify: an image block of 10mm in length is cut out every 10mm from the upper left corner of the boundingbox of the endless belt until the lower right corner is finished, an image of 10mm in length is cut out at the left or right side of the starting point as a first second image block, then a second image block is cut out at the starting point at 10mm along the direction of cutting out the first second image block, and a plurality of second image blocks are cut out successively without intervals. Wherein the boundingbox is a bounding box representing the largest cube outside an object. Preferably, a first second image block is taken at a first location (10, 10, 10-41, 41, 41), the size of the image block being 32 x 32, and then a second image block is taken from a next location (42, 42, 42-74, 74, 74) of the first second image block until the end location is taken.

Step S402: and respectively inputting the plurality of second image blocks into the blood vessel segmentation network to obtain a plurality of blood vessel segmentation images.

Specifically, the intercepted second image blocks are respectively input into a blood vessel segmentation network, and each second image block correspondingly obtains a blood vessel segmentation image.

Step S403: and splicing the plurality of blood vessel segmentation images to obtain an initial blood vessel image.

Specifically, the obtained multiple blood vessel segmentation images are spliced in sequence to obtain an initial blood vessel image. The second image block includes a position code. Specifically, the annular band is divided, and when a plurality of second image blocks are cut out at a starting position in a preset step length in sequence without intervals, a position code is allocated to each second image block, and the position code can identify the position of the second image block in the annular band. Acquiring the position code of each blood vessel segmentation image in an annular band according to the position code of each second image block; and filling each blood vessel segmentation image to a position corresponding to the position code according to the position code of the blood vessel segmentation image, so as to obtain an initial blood vessel image. More specifically, since each second image block includes a position code, after the second image block is input into the vessel segmentation network, the position codes of the second image block are allocated to the corresponding vessel segmentation images, and all the vessel segmentation images are filled to positions corresponding to the position codes according to the positions identified by the position codes, so as to obtain the initial vessel image.

In one embodiment, the input of the vessel segmentation network may not be limited to the image blocks of the annular band cut. The image blocks cut by the annular band can be used as a first input channel, the tissue organ images in the position range of the image blocks cut by the annular band can be used as a second input channel, and the images of the first channel and the second channel are input into the blood vessel segmentation network.

According to the method for acquiring the initial blood vessel image, the blood vessel image corresponding to the tissue organ is acquired on the basis of tissue organ segmentation. Because the structure of the blood vessel is complex and changeable, and the tissue organs are stable, firstly, a deep learning model is adopted, the image is firstly divided into images of all the tissue organs, reliable annular positioning is provided for the division of the blood vessel through the images of the tissue organs, a large number of areas with smaller existence probability of the blood vessel are removed, the training efficiency and the convergence speed of a blood vessel division network can be improved, and meanwhile, the possible errors of the blood vessel division can be greatly reduced.

In one embodiment, a method for acquiring a final blood vessel image is provided, and fig. 5 is a schematic flow chart of the method for acquiring a final blood vessel image in one embodiment, as shown in fig. 5, and the flow chart includes the following steps:

Step S501: and inputting the first image block into a vessel tracking network to perform end point tracking to obtain an end point tracking image.

Specifically, the end point tracking image is an image obtained by tracking and extending the end point based on the end point of the first image block.

Step S502: and extracting the connected domain according to the end point tracking image.

Specifically, the connected domain generally refers to an image region composed of foreground pixels having the same pixel value and adjacent in position in the image. In the end tracking image, the area is formed by the pixel points where the blood vessels are located. Namely, the connected domain where the blood vessel is located is extracted from the end point tracking image.

Step S503: and splicing the connected domain to the initial blood vessel image, and counting the splicing times.

Specifically, the connected domain obtained in the above step is spliced to the corresponding end point of the initial blood vessel image, and the number of times of splicing the connected domain at the end point is counted.

Step S504: and if the splicing times are greater than or equal to the preset times, obtaining a final blood vessel image.

Step S505: and if the splicing times are smaller than the preset times, extracting the first image blocks in the preset range around the blood vessel center line end point again to carry out end point tracking until the splicing times are larger than or equal to the preset times, and obtaining a final blood vessel image.

Specifically, the preset number of times may be 3 to 20 times. And (3) taking the preset times as 5 times for illustration, and if the times of splicing the connected domain at the end point is greater than or equal to 5 times, taking the image obtained after the last splicing is completed as a final blood vessel image. If the number of times of splicing the connected domain at the end point is less than 5 times, acquiring the end point of the blood vessel center line again based on the blood vessel image which is spliced at the last time, selecting an image in a preset range around the end point as a first image block, inputting the first image block into a blood vessel tracking network to carry out end point tracking, and obtaining an end point tracking image; extracting a connected domain according to the endpoint tracking image; splicing the connected domain to the blood vessel image which is spliced for the last time, and counting the splicing times of the end points again until the splicing times are more than or equal to 5 times, so as to obtain a final blood vessel image.

In one embodiment, the input to the vessel tracking network is not limited to a first image block within a predetermined range around the endpoint. The first image block may be used as a first input channel, and any one of the tissue organ image block and the initial blood vessel image block within the position range of the first image block may be used as a second input channel, and both the first input channel and the second input channel may be input to the blood vessel tracking network. The first image block can be used as a first input channel, the tissue organ image block in the position range of the first image block can be used as a second input channel, the initial blood vessel image block in the position range of the first image block can be used as a third channel, and the first input channel, the second input channel and the third channel are simultaneously input into the blood vessel tracking network.

In one embodiment, the vessel tracking method may perform the vessel direction extraction according to the initial segmentation result of the vessel before extracting the vessel center line endpoint image block, and then extract the vessel according to the main direction, so as to ensure that the vessel direction is parallel to the x direction (or the y direction and the z direction) of the extracted image block, thereby achieving the effects of simplifying the deep learning and improving the segmentation performance.

According to the method for acquiring the final blood vessel image, the segmentation model network based on the whole image is likely to form fracture, so that the clinical practical use is affected. The segmentation task is greatly simplified based on the local image vessel tracking model network, and only the reinforced vessels and the non-reinforced vessels need to be distinguished. The image to be detected is CTA data, namely data which is marked with contrast agent; the coronary vessels are now all brighter vessels. The vessel segmentation network focuses on the segmentation of the whole vessel, i.e. the separation of the coronary vessel from other tissue. Other tissues include other non-coronary vessels and other tissue organs including, for example, the heart, lungs, ribs, etc. The task of the vessel tracking network is similar to that of the vessel segmentation network, namely the coronary vessel is separated from other tissues, the scope of action is greatly reduced, the tissue is a fracture part tissue of an initial vessel image, the tissue content and the tissue complexity to be identified are greatly simplified, and only the vessel tissue and the very adjacent tissue are generally separated. Wherein the very adjacent tissue is typically only non-reinforced blood vessels, parts of the pulmonary tissue and parts of the cardiac tissue. Therefore, the segmentation of the broken blood vessel can be effectively connected, and the integrity of the segmentation of the blood vessel is improved. And by setting the preset splicing times, the segmentation efficiency can be improved on the premise of ensuring the accuracy of segmentation blood vessel images.

In one embodiment, a method for acquiring a final blood vessel image is provided, and fig. 6 is a schematic flow chart of a method for acquiring a final blood vessel image in another embodiment, as shown in fig. 6, the flow chart includes the following steps:

step S601: and inputting the first image block into a vessel tracking network to perform end point tracking to obtain an end point tracking image.

Step S602: and extracting the connected domain according to the end point tracking image.

Step S603: and splicing the connected domain to the initial blood vessel image, and acquiring a first length of a blood vessel center line in the end point tracking image and a second length of the blood vessel center line in a first image block corresponding to the end point tracking image.

Specifically, the connected domain obtained in the step is spliced to the corresponding end point of the initial blood vessel image, a first length of the blood vessel center line in the end point tracking image corresponding to the spliced connected domain is obtained, and a second length of the blood vessel center line in the first image block corresponding to the end point tracking image is obtained.

Step S604: and the first length and the second length are subjected to difference to obtain an increasing distance.

Step S605: and if the growing distance is smaller than the preset distance, obtaining a final blood vessel image.

Step S606: and if the growing distance is greater than or equal to the preset distance, extracting the first image blocks in the preset range around the center line end point of the blood vessel again to track the blood vessel until the growing distance is smaller than the preset distance, and obtaining a final blood vessel image.

Specifically, taking a preset distance of 1mm as an example for distance illustration, if the increasing distance is less than 1mm, the image obtained after the last splicing is completed is taken as a final blood vessel image. If the growing distance is greater than or equal to 1mm, acquiring the end point of the blood vessel center line again based on the blood vessel image which is spliced in the previous time, selecting an image in a preset range around the end point as a first image block, inputting the first image block into a blood vessel tracking network for end point tracking, and obtaining an end point tracking image; extracting a connected domain according to the endpoint tracking image; and splicing the connected domain to the blood vessel image which is spliced in the previous time, and calculating the growing distance again until the growing distance is smaller than 1mm, so as to obtain a final blood vessel image.

According to the method for acquiring the final blood vessel image, the segmentation model network based on the whole image is likely to form fracture, so that the clinical practical use is affected. The segmentation task is greatly simplified on the basis of the local image vessel tracking model network, and only the reinforced vessels and the non-reinforced vessels are needed to be distinguished, so that the segmentation of the broken vessels can be effectively connected, the integrity of the vessel segmentation is improved, namely, the vessel segmentation network and the vessel tracking network are in serial relation, the vessel segmentation is advanced, and then the vessel tracking is carried out. And by setting the preset distance, when the increasing distance is smaller than the preset distance, the increasing distance of the current endpoint is indicated to reach the limit soon. The tracking condition of the end point can be accurately judged by setting a proper preset distance, and the vascular tracking is timely terminated, so that the tracking effect of the end point can be ensured, the tracking time can be saved, and the segmentation efficiency can be improved.

In one embodiment, the number of splices and the distance of increase can be detected simultaneously, and when any one condition is satisfied, the end point tracking is finished, and a final blood vessel image is obtained.

In one embodiment, the distance description is performed with the medical imaging device as CT and the blood vessel as coronary. As shown in fig. 7, a process flow of coronary artery segmentation is provided, firstly, an image to be detected obtained by CT scanning is obtained, the image to be detected is firstly subjected to heart segmentation based on deep learning, an annular band is generated by calculating a distance field on the basis of the heart segmentation, the annular band is subjected to coronary artery rough segmentation based on deep learning to obtain an initial coronary artery image, then, on the basis of the initial coronary artery image, coronary artery tracking based on deep learning is performed, and the stability of the heart segmentation, the integrity of the initial coronary artery segmentation and the locality of the coronary artery tracking are combined, so that a more accurate and stable coronary artery segmentation effect is obtained.

More specifically, an object to be scanned is scanned by a Computed Tomography (CT) device, original CT data is obtained, and an image to be measured is generated. Inputting the image to be detected into a heart segmentation network for heart segmentation to obtain a heart segmentation image shown in fig. 8, wherein the heart segmentation image comprises the following steps: the aorta, right atrium, right ventricle, left atrium, and left ventricle. The distance statistics of the heart and coronary artery ranges is carried out on 800 heart segmentation images, so that a position relation statistical table of coronary artery and heart segmentation results is obtained, and the position relation statistical table is as follows:

TABLE 1 statistics of the position relationship between coronary artery and cardiac segmentation results

According to table 1, we respectively calculate the distance fields from the foreground and the background on the heart segmentation result, and acquire the possible regions of the coronary artery, namely, the annular band, as shown in fig. 9, wherein the black annular region is the possible region of the coronary artery, namely, the annular band, the blood vessels with various colors in the black annular band are the branches of the coronary artery, and the other regions are the regions with very small occurrence probability of the coronary artery.

Cutting effective image blocks in the annular band, namely cutting a plurality of image blocks continuously and at intervals in sequence with a fixed step length, and sending the image blocks into a coronary artery segmentation network to obtain an initial coronary artery segmentation result. As shown in fig. 10, fig. 10 (a) shows an initial coronary artery segmentation effect, fig. 10 (b) shows a coronary artery tracking effect, and fig. 10 (c) shows a coronary artery segmentation gold standard mark. As can be seen from fig. 10, a plurality of areas with segmentation breaks appear in the image of the initial coronary artery segmentation effect, and on this basis, the coronary artery tracking is performed on the initial coronary artery segmentation effect through the coronary artery tracking network, so as to obtain an accurate coronary artery segmentation result.

In one embodiment, a coronary tracking process and effect is performed, as shown in FIG. 11, where FIG. 11 (a) is an input image block of a coronary segmentation network; FIG. 11 (b) is an initial coronary segmentation result output by the coronary segmentation network; FIG. 11 (c) is a coronary centerline extracted based on FIG. 11 (b); FIG. 11 (d) is an input coronary endpoint image block of the coronary tracking network; FIG. 11 (e) is a coronary tracking image output by the coronary tracking network; fig. 11 (f) is an image obtained by stitching the coronary tracking image to the initial coronary segmentation result. Fig. 11 (a) is a three-view illustration of an image block, wherein one slice contains all of the coronary arteries because the selected slice size is larger. Fig. 11 (b) is an initial coronary image, fig. 11 (f) is a final coronary image obtained after coronary tracking, and fig. 11 (f) is coronary-incremented relative to fig. 11 (b) at the end point position of the coronary artery and at the end point position of the fracture. First, the coronary centerline is extracted from the initial coronary artery segmentation result output from the coronary artery segmentation network, and the extraction result of the centerline is shown in fig. 11 (c). The end points, i.e., the points of the neighborhood of only 1, are extracted on the coronary centerline, as in the intersection position of the broken lines in fig. 11 (c). The image blocks are cut with each end point as the center, as shown in fig. 11 (d), and the image blocks are sent into a trained coronary tracking network, so as to obtain a coronary tracking image, as shown in fig. 11 (e). And extracting a connected domain containing a midpoint from the coronary tracking image, and then splicing the connected domain to an initial coronary segmentation result to obtain an image shown in fig. 11 (f). The connected domain generally refers to an image area formed by foreground pixel points which have the same pixel value and are adjacent in position in the image. In the end tracking image, the area is formed by the pixel points where the blood vessels are located. The midpoint is the midpoint of the blood vessel. Finally judging whether the coronary artery segmentation result needs to continuously track the end points of the coronary artery, and stopping tracking if any condition is met: in the first condition, the tracking times are greater than or equal to N1, and the selectable range of N1 is 3-20 times; and secondly, extracting a central line from the obtained connected domain, wherein the central line is compared with an original segmentation result or central line, the growth distance of the central line is smaller than N2, and the N2 can be 1/2 of the size of the segmented image. If any of the above conditions is met, a final coronary image is obtained; if not, continuing to track the coronary artery, and finally meeting the conditions to obtain a final coronary artery image. The above-described embodiments are coronary artery segmentation techniques based on cardiac segmentation results. Compared with the complex and changeable coronary vascular structure, the atrial and ventricular structures of the heart are stable. Therefore, each chamber of the heart is firstly segmented by adopting a deep learning model, so that very reliable annular positioning of the coronary range is provided for coronary segmentation, and a large number of areas where non-coronary arteries possibly exist are removed. Not only can the training efficiency and the convergence speed of the coronary artery segmentation model be improved, but also the possible errors of the coronary artery segmentation can be greatly reduced.

The above embodiment is a method of combining a blood vessel segmentation network based on a whole image and a blood vessel tracking network based on a partial image. The blood vessel segmentation network of the whole image has a larger image field of view, and can effectively distinguish coronary arteries from veins adjacent to the veins enhanced by the developer by utilizing a heart structure, so that false positives (namely blood vessels outside the coronary arteries are segmented) are reduced. However, due to the complexity and high variability of coronary vessel structures, long and thin vessels and limited image resolution, the segmentation results output based on the vessel segmentation network tend to form cracks, which affect clinical practical use. The segmentation task is greatly simplified based on the local image vessel tracking network, and only the reinforced vessels and the non-reinforced vessels need to be distinguished, so that the segmentation of the broken vessels can be effectively connected, and the integrity of the segmentation of the vessels is improved. The method for combining the blood vessel segmentation network based on the whole image and the blood vessel tracking network based on the partial image can more accurately and completely segment the coronary artery.

Testing the above embodiment, collecting 189 cases of CTA chest data which are not used for network training, testing, and requesting a doctor to label coronary vessels (doctor label is strictly according to SCCT 18-segment branch standard, SCCT: american cardiovascular CT society), meanwhile, segmenting coronary vessels by using the coronary segmentation method of the above embodiment, and comparing results, wherein after passing through a vessel segmentation network, the comparison result is shown in figure 12, the comparison of the results of the coronary segmentation method of the above embodiment and the manually labeled segmentation results is shown in the first column in figure 12, the comparison of the blood vessel segmentation results after adding vessel tracking and the manually labeled results is shown in the second column in the following figure, wherein DICE is a measure based on the whole segmentation result, OV ^[4] The method is a segmentation measurement method based on the blood vessel center line, because the coronary blood vessel is of a slender tubular structure, branches of different areas have different thickness radii which are up to several times, and compared with DICE coefficients based on the whole blood vessel segmentation result, the method can reflect more sensitively and accurately the more frequent errors such as fracture in blood vessel segmentation by adopting the measurement standard based on the blood vessel center line. In addition, the judgment of the clinical imaging doctor on the coronary lesions depends on the visualization technology such as CPR, S-CPR or cross section and the like extracted based on the coronary central line, so that the segmentation measurement index OV based on the central line can more accurately reflect the requirements of actual clinical application.

The number in which the minimum distance of the centerline point of the gold standard from the centerline of the segmented result is smaller than the radius of the segmented result is referred to as TPR _OV Is larger than the dividing junctionThe number of points of the radius of the fruit is called FN _OV The number of points of the center line point of the segmentation result from the center line of the golden standard result by which the radius of the golden standard is smaller than the center line of the golden standard is called TPM _OV The number of points greater than the radius of the gold standard is called FP _OV 。

As can be seen from fig. 12: 1. after vessel tracking, overall performance (dic and OV) is close; 2. after the blood vessel tracking, the recovery value is greatly improved, and the repairing effect of the blood vessel tracking on the broken blood vessel is reflected; 3. after vessel tracking, the precision value drops because the vessel tracking network traces back to deeper and finer vessel ends or coronary branches not within the SCCT18 segment, such as the segmentation results shown in fig. 13, fig. 13 is a graph of the coronary segmentation results versus the gold standard for one embodiment; the left image is the coronary segmentation result of this embodiment; the right image is marked by gold standard, different line segments represent different coronary branches, the segmentation part indicated by the arrow is OM3, and the segmentation part is not in the SCCT18 segment range, but belongs to coronary vessels, and has diagnostic value in clinic.

It should be understood that, although the steps in the flowcharts of fig. 1, 4-6 are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps of FIGS. 1, 4-6 may include multiple sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor does the order in which the sub-steps or stages are performed necessarily occur in sequence, but may be performed alternately or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

In one embodiment, as shown in fig. 14, there is provided a blood vessel segmentation device including: an acquisition module 100, a tissue organ segmentation module 200, a distance field calculation module 300, a ring belt calculation module 400, a vessel segmentation module 500, an endpoint extraction module 600, and an endpoint tracking module 700, wherein:

an acquisition module 100, configured to acquire an image to be measured;

the tissue and organ segmentation module 200 is used for inputting the image to be detected into a tissue and organ segmentation network to obtain a tissue and organ image;

A distance field calculation module 300, configured to obtain a distance field of a blood vessel and a tissue organ corresponding to the blood vessel according to the tissue organ image;

an annular band calculation module 400 for determining an annular band containing the blood vessel from the distance field and tissue organ images;

the blood vessel segmentation module 500 is used for inputting the annular band into the blood vessel segmentation network to obtain an initial blood vessel image;

the endpoint extraction module 600 is configured to extract, according to the initial blood vessel image, a first image block within a preset range around a blood vessel center line endpoint;

the end point tracking module 700 is configured to input the first image block into a vessel tracking network for end point tracking, so as to obtain a final vessel image.

The blood vessel segmentation module 500 is further configured to segment the annular band into a plurality of second image blocks according to a preset size; respectively inputting a plurality of second image blocks into the blood vessel segmentation network to obtain a plurality of blood vessel segmentation images; and splicing the plurality of blood vessel segmentation images to obtain an initial blood vessel image.

The blood vessel segmentation module 500 is further configured to obtain a position code of each blood vessel segmentation image in an annular band according to the position code of each second image block; and filling each blood vessel segmentation image to a position corresponding to the position code according to the position code of the blood vessel segmentation image, so as to obtain an initial blood vessel image.

The endpoint extraction module 600 is further configured to extract a vessel centerline according to the initial vessel image; searching the end point of the blood vessel center line according to the blood vessel center line; and taking the image in a preset range around the center line end point of the blood vessel as a first image block.

The endpoint tracking module 700 is further configured to input the first image block into a vessel tracking network for endpoint tracking, so as to obtain an endpoint tracking image; extracting a connected domain according to the endpoint tracking image; and splicing the connected domain to the initial blood vessel image to obtain a final blood vessel image.

The endpoint tracking module 700 is further configured to splice the connected domain to an initial vessel image, and count the number of splices; if the splicing times are greater than or equal to the preset times, obtaining a final blood vessel image; and if the splicing times are smaller than the preset times, extracting the first image blocks in the preset range around the blood vessel center line end point again to carry out end point tracking until the splicing times are larger than or equal to the preset times, and obtaining a final blood vessel image.

The end point tracking module 700 is further configured to stitch the connected domain to the initial blood vessel image, and obtain a first length of a blood vessel center line in the end point tracking image and a second length of a blood vessel center line in a first image block corresponding to the end point tracking image; the first length and the second length are subjected to difference to obtain an increasing distance; if the growing distance is smaller than the preset distance, obtaining a final blood vessel image; and if the growing distance is greater than or equal to the preset distance, extracting the first image blocks in the preset range around the center line end point of the blood vessel again to track the blood vessel until the growing distance is smaller than the preset distance, and obtaining a final blood vessel image.

For specific limitations of the vessel segmentation device, reference may be made to the limitations of the vessel segmentation method described above, and no further description is given here. The various modules in the vessel segmentation apparatus described above may be implemented in whole or in part by software, hardware, and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.

In one embodiment, a computer device is provided, which may be a terminal, and an internal structure diagram thereof may be as shown in fig. 15. The computer device includes a processor, a memory, a network interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a vessel segmentation method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, can also be keys, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.

It will be appreciated by those skilled in the art that the structure shown in fig. 15 is merely a block diagram of a portion of the structure associated with the present application and is not limiting of the computer device to which the present application is applied, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is provided comprising a memory and a processor, the memory having stored therein a computer program, the processor when executing the computer program performing the steps of:

acquiring an image to be detected; inputting the image to be detected into a tissue organ segmentation network to obtain a tissue organ image; acquiring a distance field of a blood vessel and a tissue organ corresponding to the blood vessel according to the tissue organ image; determining an annular band containing the blood vessel from the distance field and tissue organ images; inputting the annular band into the blood vessel segmentation network to obtain an initial blood vessel image; extracting a first image block in a preset range around a blood vessel center line endpoint according to the initial blood vessel image; and inputting the first image block into a vessel tracking network for end point tracking to obtain a final vessel image.

In one embodiment, the processor when executing the computer program further performs the steps of:

dividing the annular belt into a plurality of second image blocks according to a preset size; respectively inputting a plurality of second image blocks into the blood vessel segmentation network to obtain a plurality of blood vessel segmentation images; and splicing the plurality of blood vessel segmentation images to obtain an initial blood vessel image.

acquiring the position code of each blood vessel segmentation image in an annular band according to the position code of each second image block; and filling each blood vessel segmentation image to a position corresponding to the position code according to the position code of the blood vessel segmentation image, so as to obtain an initial blood vessel image.

extracting a blood vessel center line according to the initial blood vessel image; searching the end point of the blood vessel center line according to the blood vessel center line; and taking the image in a preset range around the center line end point of the blood vessel as a first image block.

Inputting the first image block into a vessel tracking network to perform end point tracking to obtain an end point tracking image; extracting a connected domain according to the endpoint tracking image; and splicing the connected domain to the initial blood vessel image to obtain a final blood vessel image.

splicing the connected domain to an initial blood vessel image, and counting the splicing times; if the splicing times are greater than or equal to the preset times, obtaining a final blood vessel image; and if the splicing times are smaller than the preset times, extracting the first image blocks in the preset range around the blood vessel center line end point again to carry out end point tracking until the splicing times are larger than or equal to the preset times, and obtaining a final blood vessel image.

splicing the connected domain to the initial blood vessel image, and acquiring a first length of a blood vessel center line in an end point tracking image and a second length of the blood vessel center line in a first image block corresponding to the end point tracking image; the first length and the second length are subjected to difference to obtain an increasing distance; if the growing distance is smaller than the preset distance, obtaining a final blood vessel image; and if the growing distance is greater than or equal to the preset distance, extracting the first image blocks in the preset range around the center line end point of the blood vessel again to track the blood vessel until the growing distance is smaller than the preset distance, and obtaining a final blood vessel image.

In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of:

In one embodiment, the computer program when executed by the processor further performs the steps of:

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the various embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The above examples merely represent a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the invention. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application is to be determined by the claims appended hereto.

Claims

1. A method of vessel segmentation, comprising:

acquiring an image to be detected;

inputting the image to be detected into a tissue organ segmentation network to obtain a tissue organ image;

acquiring a distance field of a blood vessel and a tissue organ corresponding to the blood vessel according to the tissue organ image; the distance field comprises a distance of a blood vessel to a tissue organ wall;

determining an annular band containing the blood vessel from the distance field and tissue organ images;

Inputting the annular band into the blood vessel segmentation network to obtain an initial blood vessel image;

extracting a first image block in a preset range around a blood vessel center line endpoint according to the initial blood vessel image;

and inputting the first image block into a vessel tracking network for end point tracking to obtain a final vessel image.

2. The vessel segmentation method as set forth in claim 1, wherein inputting the annular band into the vessel segmentation network to obtain an initial vessel image comprises:

dividing the annular belt into a plurality of second image blocks according to a preset size;

respectively inputting a plurality of second image blocks into the blood vessel segmentation network to obtain a plurality of blood vessel segmentation images;

and splicing the plurality of blood vessel segmentation images to obtain the initial blood vessel image.

3. The vessel segmentation method as set forth in claim 2, wherein the stitching the plurality of vessel segmentation images to obtain the initial vessel image comprises: the second image block includes a position code;

acquiring the position code of each blood vessel segmentation image in an annular band according to the position code of each second image block;

and filling each blood vessel segmentation image to a position corresponding to the position code according to the position code of the blood vessel segmentation image, so as to obtain the initial blood vessel image.

4. The vessel segmentation method as set forth in claim 1, wherein extracting a first image block within a preset range around a vessel centerline endpoint from the initial vessel image comprises:

extracting a blood vessel center line according to the initial blood vessel image;

searching the end point of the blood vessel center line according to the blood vessel center line;

and taking the image in a preset range around the center line end point of the blood vessel as a first image block.

5. The method of claim 1, wherein inputting the first image block into a vessel tracking network for endpoint tracking, obtaining a final vessel image comprises:

inputting the first image block into a vessel tracking network to perform end point tracking to obtain an end point tracking image;

extracting a connected domain according to the endpoint tracking image;

and splicing the connected domain to the initial blood vessel image to obtain the final blood vessel image.

6. The vessel segmentation method as set forth in claim 5, wherein the stitching the connected domain to the initial vessel image to obtain the final vessel image comprises:

splicing the connected domain to an initial blood vessel image, and counting the splicing times;

If the splicing times are greater than or equal to the preset times, obtaining a final blood vessel image;

and if the splicing times are smaller than the preset times, extracting the first image blocks in the preset range around the blood vessel center line end point again to carry out end point tracking until the splicing times are larger than or equal to the preset times, and obtaining the final blood vessel image.

7. The vessel segmentation method as set forth in claim 5, wherein the stitching the connected domain to the initial vessel image to obtain a final vessel image comprises:

splicing the connected domain to the initial blood vessel image, and acquiring a first length of a blood vessel center line in an end point tracking image and a second length of the blood vessel center line in a first image block corresponding to the end point tracking image;

the first length and the second length are subjected to difference to obtain an increasing distance;

if the growing distance is smaller than the preset distance, obtaining a final blood vessel image;

and if the growing distance is greater than or equal to the preset distance, extracting the first image blocks in the preset range around the center line end point of the blood vessel again to track the blood vessel until the growing distance is smaller than the preset distance, and obtaining the final blood vessel image.

8. A vascular segmentation device, comprising:

the acquisition module is used for acquiring the image to be detected;

the tissue organ segmentation module is used for inputting the image to be detected into a tissue organ segmentation network to obtain a tissue organ image;

the distance field calculation module is used for acquiring a distance field of a blood vessel and a tissue organ corresponding to the blood vessel according to the tissue organ image; the distance field comprises a distance of a blood vessel to a tissue organ wall;

an annular band calculation module for determining an annular band containing the blood vessel based on the distance field and tissue organ images;

the blood vessel segmentation module is used for inputting the annular belt into the blood vessel segmentation network to obtain an initial blood vessel image;

the endpoint extraction module is used for extracting a first image block in a preset range around the endpoint of the blood vessel center line according to the initial blood vessel image;

and the end point tracking module is used for inputting the first image block into a blood vessel tracking network for end point tracking to obtain a final blood vessel image.

9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the vessel segmentation method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, on which a computer program is stored, characterized in that the program, when executed by a processor, implements the vessel segmentation method as claimed in any one of claims 1 to 7.