Coronary vessel segmentation method, device, equipment, storage medium and program product

Abstract

The application relates to a coronary vessel segmentation method, device, equipment, storage medium and program product. The method comprises the following steps: inputting an acquired to-be-segmented image into a segmentation network for initial segmentation, obtaining an initial vessel segmentation image of a coronary vessel in the to-be-segmented image; determining an initial tube lumen density attenuation gradient (TAG) image corresponding to the coronary vessel according to the initial vessel segmentation image of the coronary vessel; inputting the to-be-segmented image and the initial TAG image into the segmentation network for target segmentation, and determining a target vessel segmentation image of the coronary vessel; wherein the image value of the coronary vessel from the first end to the end in the initial TAG image changes linearly. The method can improve the accuracy of the obtained coronary segmentation result.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular to a method, apparatus, device, storage medium, and program product for coronary artery segmentation. Background Technology

[0002] Accurate segmentation of coronary arteries is of great significance for clinical research on coronary arteries and can also provide important prior knowledge for other applications of coronary arteries. Therefore, accurate segmentation of coronary arteries is essential.

[0003] In related technologies, images of the human heart region are typically acquired, and the obtained human heart images are input into a trained segmentation network to segment the coronary arteries in the heart region, ultimately obtaining the segmentation results of the coronary arteries and achieving the segmentation of the coronary arteries.

[0004] However, the above techniques have the problem of not obtaining accurate coronary artery segmentation results. Summary of the Invention

[0005] Therefore, it is necessary to provide a coronary artery segmentation method, apparatus, device, storage medium, and program product that can improve the accuracy of the obtained coronary artery segmentation results in response to the above-mentioned technical problems.

[0006] In a first aspect, this application provides a method for coronary artery segmentation, the method comprising:

[0007] The acquired image to be segmented is input into the segmentation network for initial segmentation to obtain the initial vessel segmentation image of the coronary vessels in the image to be segmented;

[0008] Based on the initial segmentation image of the coronary artery, the initial lumen density attenuation gradient (TAG) image corresponding to the coronary artery is determined; wherein, the image values ​​of the coronary artery from the head to the tail end in the initial TAG image show a linear change.

[0009] The image to be segmented and the initial TAG image are input into the segmentation network for target segmentation to determine the target vessel segmentation image of the coronary artery.

[0010] In one embodiment, determining the initial lumen density attenuation gradient (TAG) image corresponding to the coronary artery based on the initial vessel segmentation image of the coronary artery includes:

[0011] Based on the initial segmentation image of the coronary vessels, the image values ​​of the coronary vessels in the image to be segmented are smoothed to determine the initial TAG image corresponding to the coronary vessels.

[0012] In one embodiment, the above-mentioned smoothing of image values ​​at the coronary vessels in the image to be segmented based on the initial vessel segmentation image of the coronary vessels to determine the initial TAG image corresponding to the coronary vessels includes:

[0013] Based on the initial segmentation image of the coronary arteries, the centerline of the coronary arteries is extracted to obtain multiple centerline points and the position of each centerline point.

[0014] Map the position of each centerline point onto the image to be segmented to obtain the initial image value of each centerline point at the corresponding position on the image to be segmented.

[0015] The initial image values ​​corresponding to each centerline point are smoothed to determine the initial TAG image corresponding to the coronary artery.

[0016] In one embodiment, the above-mentioned smoothing of the initial image values ​​corresponding to each centerline point to determine the initial TAG image corresponding to the coronary artery includes:

[0017] The initial image value corresponding to each centerline point and the initial image values ​​corresponding to the surrounding centerline points are averaged to obtain the target image value of each centerline point.

[0018] The initial TAG image corresponding to the coronary artery is determined based on the target image value of each centerline point.

[0019] In one embodiment, determining the initial TAG image corresponding to the coronary artery based on the target image values ​​of each centerline point includes:

[0020] The positions of each centerline point are mapped onto the initial vessel segmentation image, and a segment of coronary vessel image corresponding to the position of each centerline point is obtained on the initial vessel segmentation image.

[0021] On the initial blood vessel segmentation image, the target image values ​​of the corresponding centerline points are filled into each segment of the coronary artery image to obtain the initial TAG image.

[0022] In one embodiment, the above-mentioned inputting the image to be segmented and the initial TAG image into the segmentation network for target segmentation to determine the target vessel segmentation image of the coronary artery includes:

[0023] Perform a segmentation operation, which includes: inputting the image to be segmented and the initial TAG image into a segmentation network for target segmentation, and determining the first segmented image of the coronary artery; and determining the first TAG image corresponding to the coronary artery based on the first segmented image of the coronary artery.

[0024] The first TAG image is used as the new initial TAG image, and the above segmentation operation is performed again until the preset iteration cutoff condition is reached, so as to obtain the target vessel segmentation image of the coronary artery.

[0025] In one embodiment, the above iteration cutoff condition includes at least one of the following:

[0026] The loss between the first segmented images in two adjacent iterations satisfies the first threshold condition;

[0027] The loss between the first TAG images in two adjacent iterations satisfies the second threshold condition;

[0028] The number of iterations has reached the preset threshold.

[0029] Secondly, this application also provides a coronary artery segmentation device, the device comprising:

[0030] The initial segmentation module is used to input the acquired image to be segmented into the segmentation network for initial segmentation, and obtain the initial vessel segmentation image of the coronary vessels in the image to be segmented;

[0031] The image determination module is used to determine the initial lumen density attenuation gradient (TAG) image corresponding to the coronary artery based on the initial vessel segmentation image of the coronary artery; wherein, the image values ​​of the coronary artery from the head to the tail end in the initial TAG image show a linear change.

[0032] The target segmentation module is used to input the image to be segmented and the initial TAG image into the segmentation network for target segmentation, and to determine the target vessel segmentation image of the coronary artery.

[0033] Thirdly, this application also provides a computer device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0034] The acquired image to be segmented is input into the segmentation network for initial segmentation to obtain the initial vessel segmentation image of the coronary vessels in the image to be segmented;

[0035] Based on the initial segmentation image of the coronary artery, the initial lumen density attenuation gradient (TAG) image corresponding to the coronary artery is determined; wherein, the image values ​​of the coronary artery from the head to the tail end in the initial TAG image show a linear change.

[0036] The image to be segmented and the initial TAG image are input into the segmentation network for target segmentation to determine the target vessel segmentation image of the coronary artery.

[0037] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0038] The acquired image to be segmented is input into the segmentation network for initial segmentation to obtain the initial vessel segmentation image of the coronary vessels in the image to be segmented;

[0039] Based on the initial segmentation image of the coronary artery, the initial lumen density attenuation gradient (TAG) image corresponding to the coronary artery is determined; wherein, the image values ​​of the coronary artery from the head to the tail end in the initial TAG image show a linear change.

[0040] The image to be segmented and the initial TAG image are input into the segmentation network for target segmentation to determine the target vessel segmentation image of the coronary artery.

[0041] Fifthly, this application also provides a computer program product, which includes a computer program that, when executed by a processor, performs the following steps:

[0042] The acquired image to be segmented is input into the segmentation network for initial segmentation to obtain the initial vessel segmentation image of the coronary vessels in the image to be segmented;

[0043] Based on the initial segmentation image of the coronary artery, the initial lumen density attenuation gradient (TAG) image corresponding to the coronary artery is determined; wherein, the image values ​​of the coronary artery from the head to the tail end in the initial TAG image show a linear change.

[0044] The image to be segmented and the initial TAG image are input into the segmentation network for target segmentation to determine the target vessel segmentation image of the coronary artery.

[0045] The aforementioned coronary artery segmentation method, apparatus, device, storage medium, and program product involve inputting the acquired image to be segmented into a segmentation network for initial segmentation to obtain initial segmentation results for the coronary arteries. Based on these initial segmentation results, an initial TAG image corresponding to the coronary arteries is determined. The image to be segmented and the initial TAG image are then input into a segmentation image to perform target segmentation, determining the target segmented image for the coronary arteries. In the initial TAG image, the image values ​​of the coronary arteries from their head to their tail exhibit a linear variation. This method, by combining the image to be segmented with the TAG image showing a linear variation in image values ​​from the head to the tail of the coronary arteries for segmentation, avoids the problem of under-segmentation or over-segmentation of the coronary arteries caused by significant differences in image values ​​between the head and tail of the coronary arteries due to poor contrast agent filling during imaging. This improves the accuracy of coronary artery segmentation. Attached Figure Description

[0046] Figure 1 This is a diagram illustrating the application environment of a coronary artery segmentation method in one embodiment.

[0047] Figure 2 This is a flowchart illustrating a coronary artery segmentation method in one embodiment;

[0048] Figure 3 This is a flowchart illustrating the coronary artery segmentation method in another embodiment;

[0049] Figure 4 This is a flowchart illustrating the coronary artery segmentation method in another embodiment;

[0050] Figure 5 This is a flowchart illustrating the coronary artery segmentation method in another embodiment;

[0051] Figure 6 This is a schematic diagram of the image to be segmented and the coronary artery segmentation map in another embodiment;

[0052] Figure 7 This is a structural block diagram of a coronary artery segmentation device in one embodiment;

[0053] Figure 8 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0055] The coronary artery segmentation method provided in this application embodiment can be applied to, for example... Figure 1In the application environment shown, the scanning device 102 can communicate with the computer device 104. Specifically, the scanning device 102 scans the object to be tested, obtains scan data, and transmits the scan data to the computer device 104 for processing. The computer device 104 can reconstruct images from the scan data, and perform image segmentation, quantization, and other processing on the images. The data storage system can store the data that the computer device 104 needs to process. The data storage system can be integrated into the computer device 104 or placed in the cloud or on other network servers. The scanning device 102 can be a single-mode or multi-mode scanning device, such as a CT (Computed Tomography) scanner or a PET-CT (Positron Emission Tomography) scanner. The computer device 104 can be a terminal or a server. If it is a terminal, it can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can be smart speakers, smart TVs, smart air conditioners, smart vehicle devices, etc. Portable wearable devices can include smartwatches, smart bracelets, head-mounted devices, etc.; if it is a server, it can be implemented using a standalone server or a server cluster composed of multiple servers.

[0056] In one embodiment, such as Figure 2 As shown, a method for coronary artery segmentation is provided, which can be applied to... Figure 1 Taking a computer device as an example, the method may include the following steps:

[0057] S202, the acquired image to be segmented is input into the segmentation network for initial segmentation to obtain the initial vascular segmentation image of the coronary vessels in the image to be segmented.

[0058] The image to be segmented can be a chest image of the subject, which may include coronary arteries, and of course, images of other tissues or organs. Specifically, it can be obtained by scanning the subject's chest with a scanning device and reconstructing the image from the obtained scan data, or by obtaining a pre-stored image from the cloud or a server. The image to be segmented can be a two-dimensional image, a three-dimensional image, or an image of other dimensions. For example, the image to be segmented can be a coronary artery CTA (CT angiography) image.

[0059] After obtaining the image to be segmented, it can be input into a segmentation network for coronary artery segmentation to obtain an initial segmented image including the coronary arteries. This initial segmented image can be, for example, a mask image of the coronary arteries, which can include the positional information of various points on the coronary arteries. The coronary arteries typically include the left coronary artery and its branches, as well as the right coronary artery and its branches.

[0060] Furthermore, the segmentation network can be a neural network model, and the specific type and architecture of the neural network model are not specifically limited here. The segmentation network can be a single-level segmentation network or a multi-level cascaded segmentation network. For example, it can first use a tissue and organ segmentation network to segment tissue and organ images, and then further use tissue and organ images and blood vessel segmentation networks to segment and obtain coronary artery images.

[0061] S204. Based on the initial segmentation image of the coronary artery, determine the initial lumen density attenuation gradient (TAG) image corresponding to the coronary artery; wherein, the image values ​​of the coronary artery from the head to the tail end in the initial TAG image show a linear change.

[0062] TAG stands for Transluminal Attenuation Gradient, which is defined as the linear regression coefficient between intraluminal density attenuation and the length from the arterial opening to the distal end.

[0063] Specifically, after obtaining the initial segmented image of the coronary arteries, the parameters of the coronary artery portion can be calculated from the beginning to the end of the segmented image by combining the initial segmented image with the coronary artery portion of the image to be segmented. This calculates the image values ​​of the coronary artery portion from beginning to end as linearly varying image values, and then the initial TAG image of the coronary arteries is obtained through the calculation result. The parameters calculated here can be the location of the coronary arteries, image values ​​or grayscale values, pixel values, etc.

[0064] S206, Input the image to be segmented and the initial TAG image into the segmentation network for target segmentation to determine the target vessel segmentation image of the coronary artery.

[0065] In this step, after obtaining the image to be segmented and the initial TAG image, the image to be segmented and the initial TAG image can be used as two input channels of the segmentation network, and the segmentation network can be used to continue segmenting the coronary vessels in the image to be segmented.

[0066] Specifically, in the process of segmenting coronary vessels on the image to be segmented using a segmentation network, the image values ​​of the coronary vessels in the initial TAG image can be used as reference information. This reference, with its linearly varying coronary vessel image values, can compensate for the differences in image values ​​of normal lumens in different coronary vessel segments caused by varying contrast agent filling states in the image to be segmented. If only the original image to be segmented is used for lumen segmentation, the gold standard lumens corresponding to different coronary vessel segments may be similar, easily confusing the segmentation network and resulting in poor segmentation. In other words, this approach avoids the problem of under-segmentation or over-segmentation of coronary vessels due to large differences in image values ​​between the beginning and end of the coronary vessel caused by poor contrast agent filling during imaging. This results in accurate coronary vessel segmentation, denoted as the target vessel segmentation image. This target vessel segmentation image has fewer over-segmentations and under-segments compared to the initial vessel segmentation image, meaning it has higher segmentation accuracy.

[0067] In the aforementioned coronary artery segmentation method, the acquired image to be segmented is input into a segmentation network for initial segmentation to obtain the initial segmentation result of the coronary arteries. Based on the initial segmentation result, an initial TAG image corresponding to the coronary arteries is determined. The image to be segmented and the initial TAG image are then input into a segmentation image to perform target segmentation, determining the target segmented image of the coronary arteries. In the initial TAG image, the image values ​​of the coronary arteries from the head to the tail exhibit a linear variation. This method, by combining the image to be segmented with the TAG image showing a linear variation in image values ​​from the head to the tail of the coronary arteries for segmentation, avoids the problem of under-segmentation or over-segmentation of the coronary arteries caused by large differences in image values ​​between the head and tail of the coronary arteries due to poor contrast agent filling during imaging. This improves the accuracy of coronary artery segmentation.

[0068] The above embodiments mentioned that the initial TAG image of the coronary artery can be obtained from the initial vessel segmentation image of the coronary artery. The following embodiments will describe the specific implementation process of this process.

[0069] In another embodiment, another coronary artery segmentation method is provided. Based on the above embodiments, S204 may include the following steps:

[0070] Based on the initial segmentation image of the coronary vessels, the image values ​​of the coronary vessels in the image to be segmented are smoothed to determine the initial TAG image corresponding to the coronary vessels.

[0071] The smoothing process can include linear smoothing and nonlinear smoothing. Linear smoothing can specifically involve performing linear smoothing on the image values ​​from the head to the tail of the coronary artery to obtain a linearly changing result. Nonlinear smoothing can specifically involve performing nonlinear smoothing on the image values ​​from the head to the tail of the coronary artery, and then further performing linear transformations to obtain a linearly changing result.

[0072] Specifically, when smoothing the image values ​​of coronary vessels in the image to be segmented from the initial vessel segmentation image, please refer to [link to relevant documentation]. Figure 3 For the method steps shown, please refer to [link / reference]. Figure 3 As shown, the specific steps may include:

[0073] S302, Based on the initial segmentation image of the coronary artery, perform centerline extraction processing on the coronary artery to obtain multiple centerline points of the coronary artery and the position of each centerline point.

[0074] In this step, after obtaining the initial blood vessel segmentation image, a centerline extraction method (such as distance transformation, morphological erosion operation, etc.) can be used to extract the centerline of the coronary blood vessels in the initial blood vessel image to obtain the centerline of the coronary blood vessels.

[0075] The image to be segmented typically includes multiple coronary arteries. Centerline extraction can be performed on each coronary artery, resulting in a centerline for each vessel. Since the initial segmented image usually includes the positions of points along the coronary arteries, the positions of these points along each centerline can also be obtained. In other words, centerline extraction of coronary arteries generally yields multiple discrete points constituting each centerline. These discrete points can be used for curve fitting to obtain the centerline, and these discrete points can be designated as centerline points, thus providing the individual centerline points and their positions.

[0076] S304, map the position of each centerline point onto the image to be segmented, and obtain the initial image value of each centerline point at the corresponding position on the image to be segmented.

[0077] In this step, after obtaining the multiple centerline points and their positions that constitute each coronary artery, since the positions of the coronary arteries in the initial coronary artery segmentation image correspond to the positions of the coronary arteries in the image to be segmented, the position of each centerline point can be mapped back to the image to be segmented. This allows us to obtain the position of the corresponding point in the image to be segmented, and simultaneously obtain the image value at that position. This mapping operation is performed on the centerline points along the centerline of each coronary artery to obtain the image value at the corresponding position of each centerline point in the image to be segmented. These obtained image values ​​are recorded as the initial image values.

[0078] The image value here can be the CT value on the image. The CT value is a unit of measurement for the density of a local tissue or organ in the human body. The CT value can also be called the HU (Hounsfield Unit) value.

[0079] Of course, regarding the centerlines of each coronary artery, taking the centerline of a single coronary artery as an example, we can start from the first centerline point on the centerline and map the image values ​​of each centerline point onto the image to be segmented. Using the distance from the centerline point to the starting point as the horizontal axis (the starting point can be set according to the actual situation) and the initial image value corresponding to that centerline point as the vertical axis, we can draw a TAG curve. This process is repeated until the end of the coronary artery. Repeating the above operation for all centerlines will yield the TAG curve information for all coronary arteries.

[0080] S306, Smooth the initial image values ​​corresponding to each centerline point to determine the initial TAG image corresponding to the coronary artery.

[0081] In this step, after obtaining the initial image values ​​corresponding to each centerline on each coronary artery, the initial image values ​​corresponding to the centerline points of each coronary artery can be smoothed using interpolation algorithms, median algorithms, mean algorithms, etc., to obtain the smoothed image values ​​of each centerline point corresponding to each coronary artery, and the initial TAG image is obtained through the smoothed image values ​​of each centerline point; in addition, assuming that the above obtains the TAG curves corresponding to each centerline, then each TAG curve can be smoothed to obtain the initial TAG image.

[0082] In this embodiment, the initial TAG image is obtained by smoothing the image values ​​of the coronary vessels in the image to be segmented from the initial segmented image of the coronary vessels. This smoothing process, combined with the segmented image and the original image, makes it relatively easy to obtain the TAG image corresponding to the coronary vessels. Furthermore, the image values ​​are obtained by mapping the centerline points of the vessels in the initial segmented image to the image to be segmented, and then smoothing is performed. This location-mapping method yields more accurate image values, leading to more accurate smoothing results. Additionally, mapping using centerline points, rather than all points, results in fewer representative points, improving the efficiency of the smoothing process while maintaining accuracy.

[0083] The above embodiments briefly illustrate the smoothing process of the initial image values ​​of each centerline point to obtain the initial TAG image. The following embodiments illustrate a possible implementation of this process.

[0084] In another embodiment, a different method for coronary artery segmentation is provided, based on the above embodiments, such as... Figure 4 As shown, the above S306 may include the following steps:

[0085] S402, the initial image value corresponding to each centerline point and the initial image values ​​corresponding to the surrounding centerline points are averaged to obtain the target image value of each centerline point.

[0086] In this step, the averaging process can include direct averaging or weighted averaging. Taking the centerline of a coronary artery as an example, it can be achieved by quickly selecting each centerline point and one or more centerline points within a certain range around the corresponding TAG curve. The initial image value of each centerline point is then averaged with the initial image values ​​of the surrounding one or more centerline points. The resulting average is used as the target image value for that centerline point. This process yields the target image value for each centerline point, ultimately resulting in the target image values ​​for all centerline points along the centerline of a coronary artery. The target image value for each centerline point along the centerline of each coronary artery can be obtained in this way. The "certain range" can be any centerline point within a distance threshold from the current centerline point; the distance threshold could be, for example, the distance between two points, etc.

[0087] By calculating the TAG curve corresponding to the centerline of each coronary artery, the centerline point for averaging can be selected relatively quickly and intuitively, thereby enabling fast and accurate averaging of image values ​​and improving the efficiency of image smoothing, which in turn further enhances the efficiency of coronary artery segmentation.

[0088] S404, determine the initial TAG image corresponding to the coronary artery based on the target image value of each centerline point.

[0089] In this step, after obtaining the target image values ​​for each centerline point, as an optional embodiment, the initial TAG image can be obtained using the following steps:

[0090] Step 1: Map the position of each centerline point onto the initial blood vessel segmentation image to obtain a segment of coronary artery image corresponding to the position of each centerline point on the initial blood vessel segmentation image.

[0091] Step 2: On the initial blood vessel segmentation image, fill the corresponding centerline point target image value for each coronary artery segment to obtain the initial TAG image.

[0092] Since the centerline points here are extracted from the initial vessel segmentation image, their positions can naturally correspond to the positions of points in the initial vessel segmentation image. Taking the centerline of a coronary artery as an example, the positions of each centerline point on this centerline can be mapped to the initial vessel segmentation image, thus obtaining a corresponding point position in the initial vessel segmentation image. Therefore, a segment of the coronary artery image can be selected at that corresponding point position in the initial vessel segmentation image. Alternatively, since the positions of the centerline points correspond to their positions in the initial vessel segmentation image, and the positions in the initial vessel segmentation image correspond to their positions in the original image to be segmented... Since the location of the coronary arteries corresponds to the position of the coronary arteries, a segment of the coronary artery image can be selected at the corresponding point in the image to be segmented. For example, the coronary artery within a 1mm range above and below the corresponding point can be selected as a segment of the coronary artery image. This selected segment of the coronary artery image is the segment of the coronary artery image corresponding to the center line point. Then, the target image value corresponding to the center line point can be filled into the entire segment of the coronary artery image. That is, the image value corresponding to the segment of the coronary artery at the center line point is the same, and all points within the range of this segment of the coronary artery will obtain the same image value as the corresponding center line point.

[0093] This operation yields a segment of the coronary artery image corresponding to each centerline point on the centerline of the coronary artery, and fills each segment of the coronary artery image with the target image value corresponding to the centerline point. Similarly, by performing this operation on the centerline of all coronary arteries, the target image value corresponding to the centerline point can be filled into all segments of the coronary artery image. The final initial segmented image with the target image value filled is the initial TAG image.

[0094] In this embodiment, the initial image values ​​of each centerline point and its surrounding points are averaged to obtain the target image value, thereby obtaining the initial TAG image. Averaging multiple points ensures a linear distribution of image values ​​at each centerline point along the centerline of a coronary artery, preventing excessively large or small image values ​​from affecting subsequent coronary artery segmentation and improving the accuracy of segmentation. Furthermore, the positions of the centerline points are mapped onto the initial vessel segmentation image, and the corresponding segments of the coronary artery image are filled with the same image value to obtain the initial TAG image. This initial TAG image is a smoothed image, providing a more accurate reference image and information for coronary artery segmentation.

[0095] The above embodiments also mention that after a segmentation process, an initial TAG image is obtained from the initial blood vessel segmentation image, and the target segmentation image of the coronary blood vessel can be obtained by segmentation using the initial TAG image and the image to be segmented. The following describes a possible implementation of this segmentation process.

[0096] In another embodiment, a different method for coronary artery segmentation is provided, based on the above embodiments, such as... Figure 5 As shown, the above S206 may include the following steps:

[0097] S502, Perform a segmentation operation, which includes: inputting the image to be segmented and the initial TAG image into a segmentation network for target segmentation, and determining the first segmented image of the coronary artery; and determining the first TAG image corresponding to the coronary artery based on the first segmented image of the coronary artery.

[0098] S504, take the first TAG image as the new initial TAG image, and return to perform the above segmentation operation until the preset iteration cutoff condition is reached, to obtain the target vessel segmentation image of the coronary artery.

[0099] In this embodiment, after obtaining the initial TAG image from the initial vessel segmentation image through the above-mentioned first segmentation process, this first segmentation process is essentially a coarse segmentation process. Then, the image to be segmented and the initial TAG image can be used as two input channels of the segmentation network to segment the coronary vessels, obtaining the segmentation result of the current coronary vessels, which is denoted as the first segmented image of the coronary vessels. Subsequently, it can be combined with the above... Figures 3-4 The process continues to process the first segmented image to obtain a new TAG image corresponding to the coronary artery, which is denoted as the first TAG image.

[0100] Subsequently, the first TAG image can be used as a new initial TAG image. The image to be segmented and the new initial TAG image are then input into the segmentation network as two input channels to segment the coronary arteries, obtaining the segmentation result of the current coronary artery, which is recorded as the new first segmented image of the coronary artery. The corresponding new first TAG image is obtained through the new first segmented image, and then the above segmentation operation is iterated back and forth until the iteration cutoff condition is reached. The segmentation result of the coronary artery obtained when the iteration cutoff condition is finally reached is recorded as the target vessel segmentation image. The coronary arteries in the target vessel segmentation image are generally relatively smooth and continuous.

[0101] As an optional embodiment, the above iteration cutoff condition includes at least one of the following: the loss between the first segmented images in two adjacent iterations meets a first threshold condition; the loss between the first TAG images in two adjacent iterations meets a second threshold condition; and the number of iterations reaches a preset number threshold.

[0102] If the loss between the first segmented images in two adjacent iterations meets the first threshold condition, after each calculation of the first segmented image, the loss can be calculated between it and the previously calculated first segmented image. The calculated loss is then compared with the first threshold condition. When the first threshold condition is met, it indicates that the coronary artery segmentation result has met the requirements, and the iteration process can be terminated. The first threshold condition can be, for example, a first loss threshold. The first threshold condition is considered met when the loss between the first segmented images in two adjacent iterations is less than this first loss threshold.

[0103] For a given set of iterations, the loss between the first TAG images in two consecutive iterations must satisfy a second threshold condition. Here, the first TAG images in two consecutive iterations are essentially the initial TAG images in those two iterations. Similarly, after each calculation of the first TAG image, the loss can be calculated between it and the previously calculated first TAG image. The calculated loss is then compared to the second threshold condition. If the second threshold condition is met, it indicates that the coronary artery segmentation result has met the requirements, and the iteration process can be terminated. This second threshold condition can be, for example, a second loss threshold. The second threshold condition is considered satisfied when the loss between the first TAG images in two consecutive iterations is less than this second loss threshold.

[0104] Once the number of iterations reaches a preset threshold, which can be set according to the actual situation (e.g., 100, 1000, etc.), the number of iterations can be continuously counted during the iteration process, and the number of iterations can be compared with the threshold. When the number of iterations is greater than the threshold, it means that the segmentation result of the coronary artery has met the requirements, and the iteration process can be stopped.

[0105] In this embodiment, the coronary artery segmentation result is obtained by further segmenting the image to be segmented and the initial TAG image. Then, a new initial TAG image is obtained based on the coronary artery segmentation result, and an iterative segmentation process is performed until the iteration cutoff condition is met to obtain the target coronary artery segmentation image. This iterative process, by obtaining the final segmentation result of the coronary arteries, can achieve higher accuracy. Furthermore, by setting different iteration cutoff conditions, the iteration process can be prevented from entering an infinite loop, ensuring the smooth progress of the coronary artery segmentation process.

[0106] The following detailed embodiment illustrates the technical solution of this application. Based on the above embodiment, the method may include the following steps:

[0107] S1, input the acquired image to be segmented into the segmentation network for initial segmentation to obtain the initial blood vessel segmentation image of the coronary vessels in the image to be segmented;

[0108] S2, Based on the initial segmentation image of the coronary vessels, perform centerline extraction processing on the coronary vessels to obtain multiple centerline points of the coronary vessel centerline and the position of each centerline point;

[0109] S3, map the position of each centerline point to the image to be segmented, and obtain the initial image value of each centerline point at the corresponding position on the image to be segmented;

[0110] S4, perform mean processing on the initial image value corresponding to each centerline point and the initial image values ​​corresponding to the surrounding centerline points to obtain the target image value of each centerline point;

[0111] S5, map the position of each centerline point to the initial blood vessel segmentation image, and obtain a segment of coronary blood vessel image corresponding to the position of each centerline point on the initial blood vessel segmentation image;

[0112] S6. On the initial blood vessel segmentation image, fill the target image value of the corresponding center line point for each segment of the coronary artery image to obtain the initial TAG image.

[0113] S7, input the image to be segmented and the initial TAG image into the segmentation network to perform target segmentation and determine the first segmented image of the coronary artery;

[0114] S8, take the first segmented image as the new initial blood vessel segmentation image, return to execute S2-S7, until the preset iteration cutoff condition is reached, and obtain the target blood vessel segmentation image of the coronary artery; wherein, the above iteration cutoff condition includes at least one of the following: the loss between the first segmented images in two adjacent iterations meets the first threshold condition; the loss between the initial TAG images in two adjacent iterations meets the second threshold condition; the number of iterations reaches the preset number threshold.

[0115] Based on the above method steps, see Figure 6As shown, (1) is the original image of the coronary arteries to be segmented, and (2) is a partial TAG image obtained after applying the method steps of this embodiment (here, the partial TAG image refers to a TAG image of a certain layer). It can be seen from the two images that the coronary arteries in this partial TAG image are clearer and more detailed than the original image to be segmented, which can provide accurate segmentation results for subsequent image analysis. In addition, by segmenting the coronary arteries together with this partial TAG image and the original image to be segmented, the segmentation process can focus more on the segmentation of the coronary artery part in the TAG image, thereby speeding up the segmentation speed and improving the segmentation efficiency and accuracy.

[0116] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0117] Based on the same inventive concept, this application also provides a coronary artery segmentation device for implementing the coronary artery segmentation method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more embodiments of the coronary artery segmentation device provided below can be found in the limitations of the coronary artery segmentation method described above, and will not be repeated here.

[0118] In one embodiment, such as Figure 7 As shown, a coronary artery segmentation device is provided, comprising: an initial segmentation module 11, an image determination module 12, and a target segmentation module 13, wherein:

[0119] The initial segmentation module 11 is used to input the acquired image to be segmented into the segmentation network for initial segmentation to obtain the initial vessel segmentation image of the coronary vessels in the image to be segmented.

[0120] The image determination module 12 is used to determine the initial lumen density attenuation gradient (TAG) image corresponding to the coronary artery based on the initial vessel segmentation image of the coronary artery; wherein, the image values ​​of the coronary artery from the head to the tail end in the initial TAG image change linearly.

[0121] The target segmentation module 13 is used to input the image to be segmented and the initial TAG image into the segmentation network for target segmentation, and to determine the target vessel segmentation image of the coronary artery.

[0122] In another embodiment, another coronary artery segmentation device is provided. Based on the above embodiments, the image determination module 12 may include an image determination unit, which is used to smooth the image values ​​of the coronary artery in the image to be segmented based on the initial segmentation image of the coronary artery, and determine the initial TAG image corresponding to the coronary artery.

[0123] Optionally, the image determination unit may include:

[0124] The extraction subunit is used to extract the centerline of the coronary artery based on the initial vessel segmentation image of the coronary artery, and obtain multiple centerline points of the coronary artery centerline and the position of each centerline point.

[0125] The mapping subunit is used to map the position of each centerline point to the image to be segmented, and obtain the initial image value of each centerline point at the corresponding position on the image to be segmented.

[0126] The smoothing subunit is used to smooth the initial image values ​​corresponding to each centerline point to determine the initial TAG image corresponding to the coronary artery.

[0127] Optionally, the smoothing subunit is specifically used to perform mean processing on the initial image value corresponding to each centerline point and the initial image values ​​corresponding to the surrounding centerline points to obtain the target image value of each centerline point; and to determine the initial TAG image corresponding to the coronary artery based on the target image value of each centerline point.

[0128] Optionally, the smoothing subunit is specifically used to map the position of each centerline point onto the initial vessel segmentation image, obtain a segment of coronary vessel image corresponding to the position of each centerline point on the initial vessel segmentation image; and fill each segment of coronary vessel image with the target image value of the corresponding centerline point to obtain the initial TAG image.

[0129] In another embodiment, another coronary artery segmentation device is provided. Based on the above embodiments, the target segmentation module 13 may include:

[0130] An execution unit is used to perform a segmentation operation, which includes: inputting the image to be segmented and an initial TAG image into a segmentation network for target segmentation, determining a first segmented image of the coronary artery; and determining a first TAG image corresponding to the coronary artery based on the first segmented image of the coronary artery.

[0131] The iteration unit is used to take the first TAG image as the new initial TAG image and return to perform the above segmentation operation until the preset iteration cutoff condition is reached to obtain the target vessel segmentation image of the coronary artery.

[0132] Optionally, the above iteration cutoff conditions include at least one of the following: the loss between the first segmented images in two adjacent iterations meets a first threshold condition; the loss between the first TAG images in two adjacent iterations meets a second threshold condition; and the number of iterations reaches a preset threshold.

[0133] Each module in the aforementioned coronary artery segmentation device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0134] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 8 As shown, the computer device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a coronary artery segmentation method. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0135] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0136] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0137] The acquired image to be segmented is input into the segmentation network for initial segmentation to obtain the initial segmentation image of the coronary vessels in the image to be segmented. Based on the initial segmentation image of the coronary vessels, the initial lumen density attenuation gradient (TAG) image corresponding to the coronary vessels is determined. The image values ​​of the coronary vessels in the initial TAG image change linearly from the head to the tail. The image to be segmented and the initial TAG image are input into the segmentation network for target segmentation to determine the target segmentation image of the coronary vessels.

[0138] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0139] Based on the initial segmentation image of the coronary vessels, the image values ​​of the coronary vessels in the image to be segmented are smoothed to determine the initial TAG image corresponding to the coronary vessels.

[0140] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0141] Based on the initial segmentation image of the coronary arteries, the centerline of the coronary arteries is extracted to obtain multiple centerline points and their positions. The positions of each centerline point are then mapped onto the image to be segmented to obtain the initial image value of each centerline point at its corresponding position on the image to be segmented. The initial image value corresponding to each centerline point is then smoothed to determine the initial TAG image corresponding to the coronary arteries.

[0142] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0143] The initial image value corresponding to each centerline point and the initial image values ​​corresponding to the surrounding centerline points are averaged to obtain the target image value of each centerline point; the initial TAG image corresponding to the coronary artery is determined based on the target image value of each centerline point.

[0144] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0145] The positions of each centerline point are mapped onto the initial vessel segmentation image, and a segment of coronary artery image corresponding to the position of each centerline point is obtained on the initial vessel segmentation image. The target image value of the corresponding centerline point is filled into each segment of coronary artery image on the initial vessel segmentation image to obtain the initial TAG image.

[0146] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0147] The segmentation operation includes: inputting the image to be segmented and the initial TAG image into the segmentation network for target segmentation to determine the first segmented image of the coronary artery; determining the first TAG image corresponding to the coronary artery based on the first segmented image of the coronary artery; using the first TAG image as the new initial TAG image and returning to perform the above segmentation operation until a preset iteration cutoff condition is reached to obtain the target vessel segmentation image of the coronary artery.

[0148] In one embodiment, the above-mentioned iteration cutoff condition includes at least one of the following: the loss between the first segmented images in two adjacent iterations meets a first threshold condition; the loss between the first TAG images in two adjacent iterations meets a second threshold condition; and the number of iterations reaches a preset number threshold.

[0149] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0150] The acquired image to be segmented is input into the segmentation network for initial segmentation to obtain the initial segmentation image of the coronary vessels in the image to be segmented. Based on the initial segmentation image of the coronary vessels, the initial lumen density attenuation gradient (TAG) image corresponding to the coronary vessels is determined. The image values ​​of the coronary vessels in the initial TAG image change linearly from the head to the tail. The image to be segmented and the initial TAG image are input into the segmentation network for target segmentation to determine the target segmentation image of the coronary vessels.

[0151] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0152] Based on the initial segmentation image of the coronary vessels, the image values ​​of the coronary vessels in the image to be segmented are smoothed to determine the initial TAG image corresponding to the coronary vessels.

[0153] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0154] Based on the initial segmentation image of the coronary arteries, the centerline of the coronary arteries is extracted to obtain multiple centerline points and their positions. The positions of each centerline point are then mapped onto the image to be segmented to obtain the initial image value of each centerline point at its corresponding position on the image to be segmented. The initial image value corresponding to each centerline point is then smoothed to determine the initial TAG image corresponding to the coronary arteries.

[0155] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0156] The initial image value corresponding to each centerline point and the initial image values ​​corresponding to the surrounding centerline points are averaged to obtain the target image value of each centerline point; the initial TAG image corresponding to the coronary artery is determined based on the target image value of each centerline point.

[0157] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0158] The positions of each centerline point are mapped onto the initial vessel segmentation image, and a segment of coronary artery image corresponding to the position of each centerline point is obtained on the initial vessel segmentation image. The target image value of the corresponding centerline point is filled into each segment of coronary artery image on the initial vessel segmentation image to obtain the initial TAG image.

[0159] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0160] The segmentation operation includes: inputting the image to be segmented and the initial TAG image into the segmentation network for target segmentation to determine the first segmented image of the coronary artery; determining the first TAG image corresponding to the coronary artery based on the first segmented image of the coronary artery; using the first TAG image as the new initial TAG image and returning to perform the above segmentation operation until a preset iteration cutoff condition is reached to obtain the target vessel segmentation image of the coronary artery.

[0161] In one embodiment, the above-mentioned iteration cutoff condition includes at least one of the following: the loss between the first segmented images in two adjacent iterations meets a first threshold condition; the loss between the first TAG images in two adjacent iterations meets a second threshold condition; and the number of iterations reaches a preset number threshold.

[0162] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0163] The acquired image to be segmented is input into the segmentation network for initial segmentation to obtain the initial segmentation image of the coronary vessels in the image to be segmented. Based on the initial segmentation image of the coronary vessels, the initial lumen density attenuation gradient (TAG) image corresponding to the coronary vessels is determined. The image values ​​of the coronary vessels in the initial TAG image change linearly from the head to the tail. The image to be segmented and the initial TAG image are input into the segmentation network for target segmentation to determine the target segmentation image of the coronary vessels.

[0164] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0165] Based on the initial segmentation image of the coronary vessels, the image values ​​of the coronary vessels in the image to be segmented are smoothed to determine the initial TAG image corresponding to the coronary vessels.

[0166] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0167] Based on the initial segmentation image of the coronary arteries, the centerline of the coronary arteries is extracted to obtain multiple centerline points and their positions. The positions of each centerline point are then mapped onto the image to be segmented to obtain the initial image value of each centerline point at its corresponding position on the image to be segmented. The initial image value corresponding to each centerline point is then smoothed to determine the initial TAG image corresponding to the coronary arteries.

[0168] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0169] The initial image value corresponding to each centerline point and the initial image values ​​corresponding to the surrounding centerline points are averaged to obtain the target image value of each centerline point; the initial TAG image corresponding to the coronary artery is determined based on the target image value of each centerline point.

[0170] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0171] The positions of each centerline point are mapped onto the initial vessel segmentation image, and a segment of coronary artery image corresponding to the position of each centerline point is obtained on the initial vessel segmentation image. The target image value of the corresponding centerline point is filled into each segment of coronary artery image on the initial vessel segmentation image to obtain the initial TAG image.

[0172] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0173] The segmentation operation includes: inputting the image to be segmented and the initial TAG image into the segmentation network for target segmentation to determine the first segmented image of the coronary artery; determining the first TAG image corresponding to the coronary artery based on the first segmented image of the coronary artery; using the first TAG image as the new initial TAG image and returning to perform the above segmentation operation until a preset iteration cutoff condition is reached to obtain the target vessel segmentation image of the coronary artery.

[0174] In one embodiment, the above-mentioned iteration cutoff condition includes at least one of the following: the loss between the first segmented images in two adjacent iterations meets a first threshold condition; the loss between the first TAG images in two adjacent iterations meets a second threshold condition; and the number of iterations reaches a preset number threshold.

[0175] It should be noted that the data involved in this application (including but not limited to data used for analysis, data stored, data displayed, etc.) are all information and data authorized by the user or fully authorized by all parties.

[0176] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0177] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0178] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method of coronary vessel segmentation, characterized by, The method comprises: inputting the obtained to-be-segmented image into a segmentation network for initial segmentation to obtain an initial blood vessel segmentation image of the coronary vessel in the to-be-segmented image; calculating the image value of the coronary vessel from the first end to the end as an image value that changes linearly according to the initial blood vessel segmentation image of the coronary vessel and the coronary vessel part in the to-be-segmented image, and determining the initial TAG image corresponding to the coronary vessel according to the image value; wherein the image value of the coronary vessel from the first end to the end in the initial TAG image changes linearly; inputting the to-be-segmented image and the initial TAG image into the segmentation network for target segmentation to determine the target blood vessel segmentation image of the coronary vessel.

2. The method of claim 1, wherein, The method comprises: calculating the image value of the coronary vessel from the first end to the end as an image value that changes linearly according to the initial blood vessel segmentation image of the coronary vessel and the coronary vessel part in the to-be-segmented image, and determining the initial TAG image corresponding to the coronary vessel according to the image value; wherein the image value of the coronary vessel from the first end to the end in the initial TAG image changes linearly.

3. The method of claim 2, wherein, The method comprises: calculating the image value of the coronary vessel from the first end to the end as an image value that changes linearly according to the initial blood vessel segmentation image of the coronary vessel and the coronary vessel part in the to-be-segmented image, and determining the initial TAG image corresponding to the coronary vessel according to the image value; wherein the image value of the coronary vessel from the first end to the end in the initial TAG image changes linearly. The method comprises: extracting the center line of the coronary vessel according to the initial blood vessel segmentation image of the coronary vessel to obtain a plurality of center line points of the center line of the coronary vessel and the positions of the center line points; 4. The method of claim 3, wherein, mapping the positions of the center line points to the to-be-segmented image to obtain the initial image values of the center line points at the corresponding positions of the to-be-segmented image; smoothing the initial image values corresponding to the center line points to determine the initial TAG image corresponding to the coronary vessel. The method comprises:

5. The method of claim 4, wherein, performing mean value processing on the initial image value corresponding to each center line point and the initial image values corresponding to the center line points around the center line point to obtain the target image values of the center line points; determining the initial TAG image corresponding to the coronary vessel according to the target image values of the center line points. The method comprises: mapping the positions of the center line points to the initial blood vessel segmentation image to obtain a section of the coronary vessel image corresponding to the positions of the center line points on the initial blood vessel segmentation image; On the initial blood vessel segmentation image, each segment of the coronary vessel image is filled with the target image value of the corresponding center line point, and the initial TAG image is obtained.

6. The method according to any one of claims 1 to 5, characterized in that, The method further includes: inputting the to-be-segmented image and the initial TAG image into the segmentation network to perform target segmentation, and determining a target blood vessel segmentation image of the coronary vessel. The method further includes: performing a segmentation operation, the segmentation operation including: inputting the to-be-segmented image and the initial TAG image into the segmentation network to perform target segmentation, and determining a first segmentation image of the coronary vessel; and determining a first TAG image corresponding to the coronary vessel according to the first segmentation image of the coronary vessel. The first TAG image is taken as a new initial TAG image, and the segmentation operation is performed until a preset iteration stop condition is reached, and a target blood vessel segmentation image of the coronary vessel is obtained.

7. A coronary vessel segmentation apparatus, characterized by, The apparatus includes: An initial segmentation module configured to input the acquired to-be-segmented image into a segmentation network to perform initial segmentation, and obtain an initial blood vessel segmentation image of the coronary vessel in the to-be-segmented image. An image determination module configured to calculate image values of the coronary vessel from a first end to a second end as image values that change linearly according to the initial blood vessel segmentation image of the coronary vessel and a coronary vessel part in the to-be-segmented image, and determine an initial lumen density attenuation gradient (TAG) image corresponding to the coronary vessel according to the image values; wherein the image values of the coronary vessel from the first end to the second end in the initial TAG image change linearly. A target segmentation module configured to input the to-be-segmented image and the initial TAG image into the segmentation network to perform target segmentation, and determine a target blood vessel segmentation image of the coronary vessel.

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that, The processor implements the steps of the method of any one of claims 1 to 6 when executing the computer program.

9. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program, when executed by the processor, implements the steps of the method of any one of claims 1 to 6.

10. A computer program product comprising a computer program, characterized in that, The computer program, when executed by the processor, implements the steps of the method of any one of claims 1 to 6. The computer program, when executed by the processor, implements the steps of the method of any one of claims 1 to 6.

Images