Chamber partitioning method, apparatus, device, storage medium, and program product

Abstract

The application relates to a chamber segmentation method, device, equipment, storage medium and program product. The method comprises the following steps: determining a first mask image of a to-be-segmented tissue according to an obtained original image; wherein the to-be-segmented tissue comprises undivided chambers and intermediate tissues between the chambers; determining a second mask image of the intermediate tissues according to the original image and the first mask image; and performing connected domain division processing on the first mask image according to the second mask image to determine a third mask image; wherein the third mask image comprises divided chambers. The method can improve the accuracy of the obtained true and false chamber segmentation results.

Description

Technical Field

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

[0002] Aortic dissection refers to the separation of blood from the aortic intima into the aortic media, causing the media to separate and extend along the long axis of the aorta, forming a true lumen and a false lumen. Blood can flow between the true and false lumens of the aorta, and thrombi may adhere to the inner wall of the false lumen, forming a blood clot zone. Detecting the true and false lumens in aortic dissection provides a valuable data foundation for further image analysis.

[0003] Currently, when segmenting the true and false lumens in aortic dissection, most methods involve first obtaining an original aortic image, and then using a segmentation model composed of neural networks to segment the original aortic image to separate the true and false lumens.

[0004] However, the above techniques suffer from the problem that the segmentation results of the real cavity and the false cavity are not accurate enough. Summary of the Invention

[0005] Therefore, it is necessary to provide a chamber segmentation method, apparatus, device, storage medium, and program product that can improve the accuracy of the obtained true and false cavity segmentation results, in order to address the above-mentioned technical problems.

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

[0007] A first mask image of the tissue to be segmented is determined based on the acquired original image; wherein the tissue to be segmented includes undivided chambers and intermediate tissues between the chambers;

[0008] A second mask image of the intermediate tissue is determined based on the original image and the first mask image;

[0009] The first mask image is partitioned into connected components based on the second mask image to determine the third mask image; wherein the third mask image includes the partitioned chambers.

[0010] In one embodiment, the above-mentioned process of performing connected component partitioning on the first mask image based on the second mask image to determine the third mask image includes:

[0011] The first location information of the intermediate tissue is determined based on the second mask image;

[0012] The first mask image is divided into at least two connected components based on the first position information to obtain the intermediate mask image;

[0013] The third mask image is determined based on the divided connected components and the intermediate mask image.

[0014] In one embodiment, determining the third mask image based on the divided connected components and intermediate mask images includes:

[0015] The centerline extraction process is performed on each connected component in the intermediate mask image to obtain the first centerline of each connected component.

[0016] The third mask image is determined based on each first center line and the intermediate mask image.

[0017] In one embodiment, determining the third mask image based on each first centerline and the intermediate mask image includes:

[0018] The centerline of the tissue to be segmented in the first mask image is extracted to obtain the second centerline corresponding to the tissue to be segmented.

[0019] Calculate the degree of overlap between each first centerline and the second centerline to obtain multiple degrees of overlap;

[0020] The third mask image is determined based on multiple overlaps and intermediate mask images.

[0021] In one embodiment, each of the aforementioned chambers includes at least a first chamber and a second chamber; the determination of the third mask image based on multiple overlaps and intermediate mask images includes:

[0022] The connected region corresponding to the first center line with the highest degree of overlap is determined as the first chamber, and the connected region corresponding to the first center line with the second highest degree of overlap is determined as the second chamber.

[0023] Based on the determined first and second chambers, each connected region in the intermediate mask image is labeled with its respective chamber to obtain the third mask image.

[0024] In one embodiment, determining the second mask image of the intermediate tissue based on the original image and the first mask image includes:

[0025] An initial mask image of the intermediate tissue is determined based on the original image and the first mask image; the intermediate tissue in the initial mask image includes at least one hole;

[0026] The holes in the initial mask image are filled to determine the second mask image.

[0027] In one embodiment, the above-described filling process for holes in the initial mask image to determine the second mask image includes:

[0028] The holes in the initial mask image are filled using a first filling method and / or a second filling method to determine the second mask image;

[0029] The first filling method mentioned above includes at least a morphological opening operation for filling small holes; the second filling method mentioned above includes at least a mesh filling method for filling large holes.

[0030] In one embodiment, the method further includes:

[0031] The target chamber is determined based on the chambers divided in the third mask image;

[0032] In the third mask image, a threshold segmentation method is used to divide the connected components corresponding to the target chamber into a first region and a second region.

[0033] In one embodiment, the above-mentioned threshold segmentation method for dividing the connected domain corresponding to the target chamber into a first region and a second region includes:

[0034] Based on the location information of the target chamber in the original image, the region corresponding to the target chamber in the original image is divided into a first region and a second region by a threshold segmentation method.

[0035] The first and second regions in the original image are mapped onto the third mask image, and the first and second regions are obtained in the third mask image.

[0036] Secondly, this application also provides a chamber partitioning device, the device comprising:

[0037] The first determining module is used to determine a first mask image of the tissue to be segmented based on the acquired original image; wherein the tissue to be segmented includes undivided chambers and intermediate tissues between the chambers;

[0038] The second determining module is used to determine a second mask image of the intermediate tissue based on the original image and the first mask image;

[0039] The segmentation module is used to perform connected component partitioning on the first mask image based on the second mask image to determine the third mask image; wherein the third mask image includes the partitioned chambers.

[0040] 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:

[0041] A first mask image of the tissue to be segmented is determined based on the acquired original image; wherein the tissue to be segmented includes undivided chambers and intermediate tissues between the chambers;

[0042] A second mask image of the intermediate tissue is determined based on the original image and the first mask image;

[0043] The first mask image is partitioned into connected components based on the second mask image to determine the third mask image; wherein the third mask image includes the partitioned chambers.

[0044] 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:

[0045] A first mask image of the tissue to be segmented is determined based on the acquired original image; wherein the tissue to be segmented includes undivided chambers and intermediate tissues between the chambers;

[0046] A second mask image of the intermediate tissue is determined based on the original image and the first mask image;

[0047] The first mask image is partitioned into connected components based on the second mask image to determine the third mask image; wherein the third mask image includes the partitioned chambers.

[0048] Fifthly, this application also provides a computer program product, comprising a computer program that, when executed by a processor, performs the following steps:

[0049] A first mask image of the tissue to be segmented is determined based on the acquired original image; wherein the tissue to be segmented includes undivided chambers and intermediate tissues between the chambers;

[0050] A second mask image of the intermediate tissue is determined based on the original image and the first mask image;

[0051] The first mask image is partitioned into connected components based on the second mask image to determine the third mask image; wherein the third mask image includes the partitioned chambers.

[0052] The aforementioned chamber segmentation method, apparatus, device, storage medium, and program product determine a first mask image of the tissue to be segmented from the acquired original image. Then, based on the original image and the first mask image, which includes the undivided chambers and the intermediate tissues between the chambers, a second mask image of the intermediate tissues is determined. Finally, the first mask image is partitioned into connected components based on the second mask image to obtain a third mask image including the divided chambers. In this method, since the mask images of the intermediate tissues between the chambers can be obtained first from the mask image of the tissue to be segmented, and the divided chambers are obtained by partitioning the intermediate tissues into connected components, the chambers are relatively complete and free from breaks, resulting in more accurate segmentation results. This improves the accuracy of the obtained chamber segmentation results. Attached Figure Description

[0053] Figure 1 Example diagram of existing aortic dissection;

[0054] Figure 2 This is an internal structural diagram of a computer device in one embodiment;

[0055] Figure 3 This is a flowchart illustrating a chamber segmentation method in one embodiment;

[0056] Figure 4 Example diagram of the gold standard for intermediate tissue in another embodiment;

[0057] Figure 5 This is a flowchart illustrating the chamber segmentation method in another embodiment;

[0058] Figure 6 This is a flowchart illustrating the chamber segmentation method in another embodiment;

[0059] Figure 7 This is a flowchart illustrating the chamber segmentation method in another embodiment;

[0060] Figure 8 This is a flowchart illustrating the chamber segmentation method in another embodiment;

[0061] Figure 9 This is a flowchart illustrating the chamber segmentation method in another embodiment;

[0062] Figure 10 This is a flowchart illustrating the chamber segmentation method in another embodiment;

[0063] Figure 11 This is a structural block diagram of a chamber division device in one embodiment. Detailed Implementation

[0064] 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.

[0065] After a tear in the aortic intima, blood flow can enter the aortic wall through the rupture, dividing the original single-lumen structure into a true lumen (TL) and a false lumen (FL). The true and false lumens may be connected or disconnected. The point of connection between the true and false lumens is the rupture site, and further ruptures may occur distally. When the true and false lumens are disconnected, they are separated by the aortic intima. Blood can flow between the true and false lumens, and thrombi may adhere to the inner wall of the false lumen, forming a blood clot zone. Currently, most methods for segmenting the true and false lumens in aortic dissection involve first obtaining an original aortic image and then using a segmentation model based on a neural network to segment the image, identifying the true and false lumens. However, see [link to relevant documentation]. Figure 1 As shown in the aortic dissection image, the image values ​​(e.g., HU values) of the true lumen and false lumen are quite similar, resulting in approximate local features. Directly using a segmentation model may fail to accurately learn the features of the true and false lumens, leading to problems such as unclear overlap of the true and false lumens and chamber breaks when using this segmentation model. In other words, the segmentation results obtained by this technique may contain breaks in the segmented true and false lumens, resulting in inaccurate segmentation results. Additionally, there are true and false lumen segmentation algorithms based on straightened images, which process the original aortic image into a SCPR image (Straightening Curved Planar Reconstruction) before performing true and false lumen segmentation. This method uses the straightened image, which may contain artifacts, potentially causing deviations from the original image and resulting in poor segmentation performance, i.e., inaccurate segmentation results. Therefore, this application provides a chamber segmentation method, apparatus, device, storage medium, and program product to solve the above-mentioned technical problems.

[0066] The chamber segmentation method provided in this application can be applied to computer equipment, which can be a terminal or a server. Taking a terminal as an example, its internal structure diagram can be as follows: Figure 2As 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 executed by the processor, the computer program implements a chamber partitioning method. The display screen can be an LCD screen or an e-ink display 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.

[0067] Those skilled in the art will understand that Figure 2 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.

[0068] In one embodiment, such as Figure 3 As shown, a chamber segmentation method is provided, which can be applied to... Figure 2 Taking a computer device as an example, the method may include the following steps:

[0069] S202, determine a first mask image of the tissue to be segmented based on the acquired original image; wherein the tissue to be segmented includes undivided chambers and intermediate tissues between chambers.

[0070] The tissue to be divided can be, for example, the aorta, carotid artery, or iliac artery, or other tissues including cavities, or tissues with multiple cavities due to a dissection tear. Taking the aorta as an example, the cavities can include a true lumen and a false lumen, and the intermediate tissue between the true lumen and the false lumen can be an intimal flap.

[0071] The original image can be obtained by real-time reconstruction of the tissue to be segmented after scanning, or it can be obtained from images of the tissue to be segmented pre-stored in the cloud or on a server. Of course, other methods of obtaining the original image are also possible, and no specific limitation is made here. The original image can be a CTA (CT angiography) image, or other types of CT (Computed Tomography) images or MRI (Magnetic Resonance Imaging) images.

[0072] After obtaining the original image of the tissue to be segmented, a segmentation model or algorithm can be used to segment the tissue in the original image, obtaining a segmented image of the tissue, i.e., the first mask image. This first mask image is a binary mask image including the tissue to be segmented and the background, where the tissue to be segmented and the background can be distinguished by two different grayscale values. Furthermore, the segmentation model used to segment the tissue can be a neural network model, such as the V-net network; the specific model type is not specifically limited here. The segmentation model can be trained using Dice loss.

[0073] It should be noted that the first mask image here only segments the tissue to be segmented, but does not divide and segment the individual cavities within the tissue; it remains a single entity. Furthermore, since the tissue to be segmented is generally large in volume or area, the first mask image obtained by segmenting the tissue in the original image using a segmentation model or algorithm has relatively high accuracy, for example, reaching over 90% segmentation accuracy. Therefore, using the first mask image as the basis for subsequent image processing will not significantly affect the accuracy of the subsequent image processing results.

[0074] S204, determine the second mask image of the intermediate tissue based on the original image and the first mask image.

[0075] In this step, the intermediate tissue can be a part of the tissue to be segmented. A segmentation model or algorithm can be used to segment the intermediate tissue from the original image and the first mask image of the tissue to be segmented, obtaining a segmented image of the intermediate tissue. The second mask image can then be obtained from the segmented image. The segmentation model for segmenting the intermediate tissue can be a neural network model, such as the V-net network; the specific model type is not limited here. The segmentation model for the intermediate tissue can be trained using Mask Dice loss. During training, a gold standard mask for the intermediate tissue can first be generated from the manually labeled gold standard masks of each cavity. The boundaries between each cavity are found based on neighborhood information, and the image values ​​of the intermediate tissue (e.g., HU values) are obtained statistically. The gold standard of the intermediate tissue is then extracted, for example... Figure 4 As shown. During the network training process using loss, since the tissue to be segmented has already been segmented in the above steps and the first mask image of the tissue to be segmented has been obtained, we can calculate the loss only for the tissue to be segmented, with the middle tissue as the foreground and the remaining tissue regions as the background, which can improve computational efficiency.

[0076] By incorporating a first mask image of the tissue to be segmented into the segmentation of the intermediate tissue, the segmentation model or algorithm can focus solely on segmenting the intermediate tissue in the original image, resulting in higher efficiency and accuracy in segmentation.

[0077] In addition, the second mask image is similar to the first mask image, and can also be a binary mask image including intermediate tissue and background, wherein the intermediate tissue and background in the second mask image can be distinguished by two different gray values.

[0078] S206, Perform connected component partitioning on the first mask image based on the second mask image to determine the third mask image; wherein the third mask image includes the partitioned chambers.

[0079] In this step, after obtaining the first mask image of the tissue to be segmented and the second mask image of the intermediate tissue, since the intermediate tissue can be a part of the tissue to be segmented, the tissue to be segmented in the first mask image can be divided into different connected regions using the intermediate tissue in the second mask image as the boundary. Different cavities can be obtained through each different connected region, that is, the cavities in the tissue to be segmented are segmented out to obtain the third mask image and the divided cavities.

[0080] In the above-described cavity segmentation method, a first mask image of the tissue to be segmented is determined using the acquired original image. A second mask image of the intermediate tissue is determined based on the original image and the first mask image, which includes the undivided cavities and the intermediate tissue between the cavities. Finally, connected component partitioning is performed on the first mask image using the second mask image to obtain a third mask image including the divided cavities. This method, by first obtaining the mask image of the intermediate tissue between the cavities from the mask image of the tissue to be segmented, and then performing connected component partitioning using the mask image of the intermediate tissue to obtain the divided cavities, results in relatively complete cavities without breaks, thus improving the accuracy of the obtained cavity segmentation results.

[0081] The above embodiments mention that the first mask image of the tissue to be segmented can be divided into connected components using the second mask image of the intermediate tissue to obtain the divided chambers. The following embodiments illustrate a specific implementation of this process.

[0082] In another embodiment, such as Figure 5 As shown, another chamber segmentation method is provided. Based on the above embodiment, S206 may include the following steps:

[0083] S302, determine the first location information of the intermediate tissue based on the second mask image.

[0084] In this step, after obtaining the second mask image of the intermediate tissue, the second mask image may include the pixel position of the intermediate tissue and the pixel value at the pixel position. The pixel position of the intermediate tissue in the second mask image can be recorded as the first position information.

[0085] S304, the first mask image is divided into at least two connected regions according to the first position information to obtain the intermediate mask image.

[0086] In this step, since the tissue to be segmented and the intermediate tissue in the first mask image and the second mask image are corresponding, the positions of the intermediate tissue in the first mask image and the second mask image are also corresponding, i.e., the same. In this way, the first position information of the intermediate tissue in the second mask image can be mapped to the first mask image to obtain the position information of the intermediate tissue in the first mask image.

[0087] Subsequently, the location information of the intermediate tissue in the first mask image can be used as the boundary for dividing the connected components. That is, using this location information as the limit, the tissue to be segmented in the first mask image can be divided into at least two connected components. The image with the divided connected components can be denoted as the intermediate mask image. Taking the aorta as the tissue to be segmented and the intima sheet as the intermediate tissue as an example, the tissue to be segmented can be divided into two connected components using the intima sheet as the boundary.

[0088] S306, determine the third mask image based on the divided connected components and intermediate mask images.

[0089] In this step, after obtaining the intermediate mask image with the connected components divided as described above, each connected component in the intermediate mask image can be directly determined as a separate cavity; alternatively, the connected components in the intermediate mask image can be post-processed, such as by edge processing, and the cavity can be determined by the post-processed connected components. In short, the final result is the image of the divided cavities, which is denoted as the third mask image.

[0090] In this embodiment, the mask image of the tissue to be segmented is divided into at least two connected components using the location information of the intermediate tissue in the mask image of the intermediate tissue. The third mask image of each cavity is determined by the intermediate mask image of the divided connected components and the divided connected components. In this way, the connected components are divided by the location information of the intermediate tissue, and then each cavity is divided. The resulting cavity division is more accurate, thereby improving the accuracy of the cavity division.

[0091] The above embodiments mention that the third mask image for each chamber can be determined by dividing each connected component and the associated intermediate mask image. The following embodiments illustrate a specific implementation of this process.

[0092] In another embodiment, such as Figure 6 As shown, another chamber segmentation method is provided. Based on the above embodiment, S306 may include the following steps:

[0093] S402, perform centerline extraction processing on each connected component in the intermediate mask image to obtain the first centerline of each connected component.

[0094] In this step, after obtaining the intermediate mask image with each connected component divided as described above, a centerline extraction method (such as distance transformation, morphological erosion, etc.) can be used to extract the centerline of each connected component, and the centerline corresponding to each connected component is denoted as the first centerline. When obtaining each first centerline, it is generally possible to obtain multiple discrete points to form the centerline, and thus obtain the position information of each discrete point on the corresponding mask image.

[0095] S404, determine the third mask image based on each first center line and the intermediate mask image.

[0096] In this step, after obtaining the first center line and intermediate mask image of each connected region, as an optional embodiment, see [link to example]. Figure 7 As shown, the third mask image can be obtained using the following steps:

[0097] S502, perform centerline extraction processing on the tissue to be segmented in the first mask image to obtain the second centerline corresponding to the tissue to be segmented.

[0098] In this step, a centerline extraction method (such as distance transformation or morphological erosion) can also be used to extract the centerline of the tissue to be segmented in the first mask image, obtaining the centerline corresponding to the tissue to be segmented, which is denoted as the second centerline. Similarly, the positional information of multiple discrete points constituting the second centerline can also be obtained here.

[0099] S504, calculate the degree of overlap between each first centerline and the second centerline to obtain multiple degrees of overlap.

[0100] In this step, after obtaining the position information of each discrete point on each first center line and each discrete point on the second center line, the position information of the discrete points on each first center line can be compared with the position information of the discrete points on the second center line to determine the number of discrete points with the same position information between each first center line and the second center line (here, points with the same position are considered to coincide). Then, the number of the same discrete points can be used as the degree of coincidence between the first center line and the second center line. In this way, the degree of coincidence corresponding to each first center line can be obtained, that is, multiple degrees of coincidence can be obtained.

[0101] S506, determine the third mask image based on multiple overlaps and intermediate mask images.

[0102] In this step, taking the aforementioned chambers, including the first chamber and the second chamber, as an example, after obtaining the overlap corresponding to each first centerline, as an optional embodiment, the third mask image can be obtained using the following steps:

[0103] The connected component corresponding to the first center line with the highest overlap is determined as the first chamber, and the connected component corresponding to the first center line with the second highest overlap is determined as the second chamber. Based on the determined first and second chambers, each connected component in the intermediate mask image is marked as a chamber to obtain the third mask image.

[0104] In other words, after obtaining multiple overlap scores, these scores can be sorted, either from high to low or low to high, to obtain a sorting result. Then, the highest overlap score (or the second highest overlap score) and the second highest overlap score (or the second highest overlap score) can be obtained from the sorting result. The connected component corresponding to the first center line of the highest overlap score can then be designated as the first chamber, and the connected component corresponding to the first center line of the second highest overlap score can be designated as the second chamber. Finally, in the intermediate mask image where each connected component has been divided, each connected component can be labeled with its corresponding chamber. Once the labeling is complete, the connected components in the intermediate mask image are divided into chambers, and this labeled mask image is the third mask image. For example, the first chamber here can be a real chamber, and the second chamber can be a fake chamber.

[0105] In this embodiment, centerlines are extracted from each connected region in the intermediate mask image, and a third mask image for dividing each cavity is determined using the obtained first centerlines of each connected region and the intermediate mask image. This method of dividing the cavity by the centerlines of the connected regions, rather than calculating the entire connected region, reduces computational load and complexity, significantly improving the efficiency of cavity division. Furthermore, dividing the cavity by the degree of overlap between the centerlines of each connected region and the centerline of the tissue to be segmented is simple and intuitive, thus improving the accuracy and efficiency of cavity division. Further, the connected region with the highest degree of overlap between its centerline and the centerline of the tissue to be segmented is designated as the first cavity, and the second highest degree is designated as the second cavity, effectively achieving the division of the first and second cavities.

[0106] The above embodiments mention that the mask image of the intermediate tissue between each chamber can be determined by the original image of the tissue to be segmented and the mask image of the tissue to be segmented. The following embodiments illustrate a specific implementation of this process.

[0107] In another embodiment, such as Figure 8 As shown, another chamber segmentation method is provided. Based on the above embodiment, S204 may include the following steps:

[0108] S602, an initial mask image of the intermediate tissue is determined based on the original image and the first mask image; the intermediate tissue in the initial mask image includes at least one hole.

[0109] In this step, as mentioned in S204 above, a segmentation model or segmentation algorithm can be used to segment the intermediate tissue in the original image and the first mask image of the tissue to be segmented, to obtain the segmented image of the intermediate tissue, i.e. the initial mask image. Since the intermediate tissue will inevitably have a breakage problem when the second mask image of the intermediate tissue is actually obtained, that is, the intermediate tissue in the initial mask image may generally have at least one hole, resulting in discontinuity and breakage of the intermediate tissue. The intermediate tissue can be processed into a continuous whole tissue without breakage through the following step S604.

[0110] S604, fill the holes in the initial mask image to determine the second mask image.

[0111] In this step, after obtaining the initial mask image, the holes in the intermediate tissue of the initial mask image can be filled. As an optional embodiment, a first filling method and / or a second filling method can be used to fill the holes in the initial mask image to determine the second mask image. The first filling method includes at least a morphological opening operation for filling small holes, and the second filling method includes at least a mesh filling method for filling large holes.

[0112] Among them, morphological opening is more suitable for filling / filling small holes, which may be small ruptures; mesh filling can convert the image into mesh format and use the triangular facet structure of mesh to fill large holes, which is more suitable for filling / filling large holes, which may be large ruptures, such as fractures in the middle tissue.

[0113] The method for determining the size of the holes mentioned here can be as follows: After obtaining the initial mask image, since the initial mask image is mostly a three-dimensional image, it includes multiple layers of two-dimensional initial mask images. Therefore, the size of the holes in the intermediate tissue can be determined by detecting the number of missing two-dimensional initial mask images in the three-dimensional image. In this way, different filling methods can be selected to fill the holes according to their different sizes, which can improve the efficiency and accuracy of hole filling.

[0114] In this embodiment, an initial mask image of intermediate tissue, including holes, is determined using the original image and a first mask image of the tissue to be segmented. The holes in the initial mask image are then filled to obtain a second mask image of the intermediate tissue. This hole-filling process makes the intermediate tissue more complete, allowing for accurate division of the cavities and improving the accuracy of the divided cavities. Furthermore, different filling methods are used to fill holes of different sizes, enabling targeted hole-filling operations and improving both the efficiency and accuracy of hole filling.

[0115] In the true and false lumens of the aorta, blood can flow between them. Thrombi may adhere to the inner wall of the false lumen, forming a blood coagulation zone. The blood coagulation zone can also play a certain auxiliary role in subsequent image analysis. Therefore, it is necessary to segment the blood coagulation zone in the actual segmentation process. The following example illustrates the process of segmenting the blood coagulation zone.

[0116] In another embodiment, such as Figure 9 As shown, another chamber segmentation method is provided. Based on the above embodiments, the method may further include the following steps:

[0117] S702, determine the target chamber based on the chambers divided in the third mask image.

[0118] In this step, taking the aorta as an example, the target chamber can be a dummy chamber. After obtaining the third mask image with the chambers divided, the target chamber can be obtained from the marked chambers in the third mask image.

[0119] S704, in the third mask image, the connected components corresponding to the target chamber are divided into a first region and a second region using a threshold segmentation method.

[0120] In this step, as an optional embodiment, the target chamber can be divided into regions in the following manner: based on the position information of the target chamber in the original image, the region corresponding to the target chamber in the original image is divided into a first region and a second region using a threshold segmentation method; the first region and the second region in the original image are mapped to a third mask image, and the divided first region and the second region are obtained in the third mask image.

[0121] In this approach, the target chamber's position corresponds in both the original image and the third mask image; that is, they are identical. Thresholding segmentation typically uses image values ​​(e.g., HU values) for segmentation, which is suitable when different regions exhibit high or low density image values. However, the third mask image is usually a binary mask image, from which image values ​​cannot be obtained. Therefore, it needs to be mapped onto the original image for thresholding segmentation. Here, the target chamber can be divided into a first region and a second region in the original image, and then these two regions can be mapped onto the third mask image for region labeling. This allows the segmented first and second regions to be obtained in the third mask image.

[0122] For example, taking a false cavity as an example, the false cavity includes a blood coagulation area. The image value of the blood coagulation area is generally low-density, while the image value of the non-blood coagulation area is generally high-density. Then, the false cavity can be divided into a blood coagulation area and a non-blood coagulation area by threshold segmentation, so that the data of these two areas can be processed quickly in the future.

[0123] In this embodiment, the target cavity is determined in the third mask image with predefined chambers, and a threshold segmentation method is used to divide the target cavity into two regions. This threshold segmentation method allows for relatively fast region division of the target cavity, improving the efficiency of region division. Furthermore, by dividing the corresponding region of the target cavity on the original image and mapping it to the third mask image, effective region division of the target cavity within the third mask image can be achieved.

[0124] Taking the aorta as an example, see [link to relevant documentation]. Figure 10 As shown, a detailed embodiment is given below to illustrate the technical solution of this application. Based on the above embodiment, the method may include the following steps:

[0125] S1, Obtain the raw image of the aorta;

[0126] S2, segment the original image of the aorta to obtain the aortic mask image;

[0127] S3. Based on the original image of the aorta and the aortic mask image, the intima patch is segmented to obtain the intima patch mask image.

[0128] S4, fill the holes in the inner membrane mask image;

[0129] S5, the aortic mask image is divided into two connected components by the intima patch mask image after filling the cavity;

[0130] S6, by performing centerline coincidence analysis between the centerline of each connected domain and the centerline of the aortic mask image, the true lumen and the false lumen are distinguished, thus enabling the two connected domains to be labeled as the true lumen and the false lumen respectively;

[0131] S7, the false lumen is divided into blood coagulation zone and non-blood coagulation zone using a threshold segmentation method;

[0132] S8, obtain mask images of the true cavity, blood coagulation area and non-blood coagulation area.

[0133] 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.

[0134] Based on the same inventive concept, this application also provides a chamber segmentation device for implementing the chamber 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 chamber segmentation device embodiments provided below can be found in the limitations of the chamber segmentation method described above, and will not be repeated here.

[0135] In one embodiment, such as Figure 11 As shown, a chamber segmentation device is provided, comprising: a first determining module 11, a second determining module 12, and a segmentation module 13, wherein:

[0136] The first determining module 11 is used to determine a first mask image of the tissue to be segmented based on the acquired original image; wherein the tissue to be segmented includes undivided chambers and intermediate tissues between the chambers;

[0137] The second determining module 12 is used to determine a second mask image of the intermediate tissue based on the original image and the first mask image;

[0138] The segmentation module 13 is used to perform connected component partitioning processing on the first mask image based on the second mask image to determine the third mask image; wherein the third mask image includes the partitioned chambers.

[0139] In another embodiment, another chamber segmentation device is provided. Based on the above embodiments, the segmentation module 13 may include:

[0140] A location information determination unit is used to determine the first location information of the intermediate tissue based on the second mask image;

[0141] A partitioning unit is used to divide the first mask image into at least two connected components according to the first position information to obtain an intermediate mask image;

[0142] The mask image determination unit is used to determine the third mask image based on the divided connected components and the intermediate mask images.

[0143] In another embodiment, a different chamber segmentation device is provided. Based on the above embodiments, the mask image determination unit may include:

[0144] The extraction sub-unit is used to extract the centerline of each connected component in the intermediate mask image to obtain the first centerline of each connected component.

[0145] The mask image determination subunit is used to determine the third mask image based on each first center line and the intermediate mask image.

[0146] Optionally, the aforementioned mask image determination subunit is specifically used to perform centerline extraction processing on the tissue to be segmented in the first mask image to obtain the second centerline corresponding to the tissue to be segmented; calculate the overlap between each first centerline and the second centerline to obtain multiple overlaps; and determine the third mask image based on the multiple overlaps and the intermediate mask image.

[0147] Optionally, each of the above-mentioned chambers includes at least a first chamber and a second chamber; the above-mentioned mask image determining subunit is specifically used to determine the connected region corresponding to the first center line with the highest overlap as the first chamber, and to determine the connected region corresponding to the first center line with the second highest overlap as the second chamber; based on the determined first chamber and second chamber, each connected region in the intermediate mask image is marked with a chamber to obtain a third mask image.

[0148] In another embodiment, another chamber segmentation device is provided. Based on the above embodiments, the second determining module 12 may include:

[0149] An initial mask image determination unit is used to determine an initial mask image of intermediate tissue based on an original image and a first mask image; the intermediate tissue in the initial mask image includes at least one hole;

[0150] A filling unit is used to fill the holes in the initial mask image to determine the second mask image.

[0151] Optionally, the aforementioned filling unit is specifically used to fill the holes in the initial mask image using a first filling method and / or a second filling method to determine the second mask image; wherein, the aforementioned first filling method includes at least a morphological opening operation for filling small-sized holes; and the aforementioned second filling method includes at least a mesh filling method for filling large-sized holes.

[0152] In another embodiment, a different chamber partitioning device is provided, which, based on the above embodiments, may further include:

[0153] The target chamber determination module is used to determine the target chamber based on the chambers divided in the third mask image;

[0154] The threshold segmentation module is used to divide the connected components corresponding to the target chamber into a first region and a second region in the third mask image using a threshold segmentation method.

[0155] Optionally, the threshold segmentation module described above may include:

[0156] The threshold segmentation unit is used to divide the region corresponding to the target chamber in the original image into a first region and a second region based on the position information of the target chamber in the original image using the threshold segmentation method.

[0157] The mapping unit is used to map the first and second regions in the original image to the third mask image, and obtain the divided first and second regions in the third mask image.

[0158] Each module in the aforementioned chamber partitioning device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0159] 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:

[0160] A first mask image of the tissue to be segmented is determined based on the acquired original image; wherein the tissue to be segmented includes undivided cavities and intermediate tissues between the cavities; a second mask image of the intermediate tissues is determined based on the original image and the first mask image; a third mask image is determined by performing connected component partitioning on the first mask image based on the second mask image; wherein the third mask image includes the divided cavities.

[0161] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0162] The first location information of the intermediate tissue is determined based on the second mask image; the first mask image is divided into at least two connected components based on the first location information to obtain an intermediate mask image; and a third mask image is determined based on the divided connected components and the intermediate mask image.

[0163] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0164] The centerline extraction process is performed on each connected component in the intermediate mask image to obtain the first centerline of each connected component; the third mask image is determined based on the first centerline and the intermediate mask image.

[0165] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0166] The centerline extraction process is performed on the tissue to be segmented in the first mask image to obtain the second centerline corresponding to the tissue to be segmented; the overlap between each first centerline and the second centerline is calculated to obtain multiple overlaps; the third mask image is determined based on the multiple overlaps and the intermediate mask image.

[0167] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0168] The connected component corresponding to the first center line with the highest overlap is determined as the first chamber, and the connected component corresponding to the first center line with the second highest overlap is determined as the second chamber. Based on the determined first and second chambers, each connected component in the intermediate mask image is marked as a chamber to obtain the third mask image.

[0169] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0170] An initial mask image for the intermediate tissue is determined based on the original image and the first mask image; the intermediate tissue in the initial mask image includes at least one hole; the hole in the initial mask image is filled to determine a second mask image.

[0171] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0172] The holes in the initial mask image are filled using a first filling method and / or a second filling method to determine the second mask image; wherein the first filling method includes at least a morphological opening operation for filling small holes; and the second filling method includes at least a mesh filling method for filling large holes.

[0173] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0174] The target cavity is determined based on the chambers divided in the third mask image; in the third mask image, the connected components corresponding to the target cavity are divided into a first region and a second region using a threshold segmentation method.

[0175] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0176] Based on the location information of the target chamber in the original image, the region corresponding to the target chamber in the original image is divided into a first region and a second region using a threshold segmentation method; the first region and the second region in the original image are mapped to the third mask image, and the divided first region and the second region are obtained in the third mask image.

[0177] 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:

[0178] A first mask image of the tissue to be segmented is determined based on the acquired original image; wherein the tissue to be segmented includes undivided cavities and intermediate tissues between the cavities; a second mask image of the intermediate tissues is determined based on the original image and the first mask image; a third mask image is determined by performing connected component partitioning on the first mask image based on the second mask image; wherein the third mask image includes the divided cavities.

[0179] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0180] The first location information of the intermediate tissue is determined based on the second mask image; the first mask image is divided into at least two connected components based on the first location information to obtain an intermediate mask image; and a third mask image is determined based on the divided connected components and the intermediate mask image.

[0181] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0182] The centerline extraction process is performed on each connected component in the intermediate mask image to obtain the first centerline of each connected component; the third mask image is determined based on the first centerline and the intermediate mask image.

[0183] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0184] The centerline extraction process is performed on the tissue to be segmented in the first mask image to obtain the second centerline corresponding to the tissue to be segmented; the overlap between each first centerline and the second centerline is calculated to obtain multiple overlaps; the third mask image is determined based on the multiple overlaps and the intermediate mask image.

[0185] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0186] The connected component corresponding to the first center line with the highest overlap is determined as the first chamber, and the connected component corresponding to the first center line with the second highest overlap is determined as the second chamber. Based on the determined first and second chambers, each connected component in the intermediate mask image is marked as a chamber to obtain the third mask image.

[0187] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0188] An initial mask image for the intermediate tissue is determined based on the original image and the first mask image; the intermediate tissue in the initial mask image includes at least one hole; the hole in the initial mask image is filled to determine a second mask image.

[0189] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0190] The holes in the initial mask image are filled using a first filling method and / or a second filling method to determine the second mask image; wherein the first filling method includes at least a morphological opening operation for filling small holes; and the second filling method includes at least a mesh filling method for filling large holes.

[0191] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0192] The target cavity is determined based on the chambers divided in the third mask image; in the third mask image, the connected components corresponding to the target cavity are divided into a first region and a second region using a threshold segmentation method.

[0193] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0194] Based on the location information of the target chamber in the original image, the region corresponding to the target chamber in the original image is divided into a first region and a second region using a threshold segmentation method; the first region and the second region in the original image are mapped to the third mask image, and the divided first region and the second region are obtained in the third mask image.

[0195] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0196] A first mask image of the tissue to be segmented is determined based on the acquired original image; wherein the tissue to be segmented includes undivided cavities and intermediate tissues between the cavities; a second mask image of the intermediate tissues is determined based on the original image and the first mask image; a third mask image is determined by performing connected component partitioning on the first mask image based on the second mask image; wherein the third mask image includes the divided cavities.

[0197] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0198] The first location information of the intermediate tissue is determined based on the second mask image; the first mask image is divided into at least two connected components based on the first location information to obtain an intermediate mask image; and a third mask image is determined based on the divided connected components and the intermediate mask image.

[0199] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0200] The centerline extraction process is performed on each connected component in the intermediate mask image to obtain the first centerline of each connected component; the third mask image is determined based on the first centerline and the intermediate mask image.

[0201] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0202] The centerline extraction process is performed on the tissue to be segmented in the first mask image to obtain the second centerline corresponding to the tissue to be segmented; the overlap between each first centerline and the second centerline is calculated to obtain multiple overlaps; the third mask image is determined based on the multiple overlaps and the intermediate mask image.

[0203] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0204] The connected component corresponding to the first center line with the highest overlap is determined as the first chamber, and the connected component corresponding to the first center line with the second highest overlap is determined as the second chamber. Based on the determined first and second chambers, each connected component in the intermediate mask image is marked as a chamber to obtain the third mask image.

[0205] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0206] An initial mask image for the intermediate tissue is determined based on the original image and the first mask image; the intermediate tissue in the initial mask image includes at least one hole; the hole in the initial mask image is filled to determine a second mask image.

[0207] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0208] The holes in the initial mask image are filled using a first filling method and / or a second filling method to determine the second mask image; wherein the first filling method includes at least a morphological opening operation for filling small holes; and the second filling method includes at least a mesh filling method for filling large holes.

[0209] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0210] The target cavity is determined based on the chambers divided in the third mask image; in the third mask image, the connected components corresponding to the target cavity are divided into a first region and a second region using a threshold segmentation method.

[0211] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0212] Based on the location information of the target chamber in the original image, the region corresponding to the target chamber in the original image is divided into a first region and a second region using a threshold segmentation method; the first region and the second region in the original image are mapped to the third mask image, and the divided first region and the second region are obtained in the third mask image.

[0213] It should be noted that the data involved in this application (including but not limited to data used for analysis, stored data, and displayed data) are all information and data authorized by the user or fully authorized by all parties.

[0214] 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.

[0215] 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.

[0216] 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 for compartment division, characterized in that, The method includes: A first mask image of the tissue to be segmented is determined based on the acquired original image; wherein the tissue to be segmented includes undivided chambers and intermediate tissues between the chambers, and each chamber includes a true chamber and a false chamber; An initial mask image for the intermediate tissue is determined based on the original image and the first mask image; the intermediate tissue in the initial mask image includes at least one hole; the hole in the initial mask image is filled to determine a second mask image; The first location information of the intermediate tissue is determined based on the second mask image; the first mask image is divided into at least two connected components based on the first location information to obtain an intermediate mask image; a third mask image is determined based on each of the divided connected components and the intermediate mask image; wherein the third mask image includes each of the divided cavities.

2. The method according to claim 1, characterized in that, Determining the third mask image based on the divided connected components and the intermediate mask image includes: The centerline extraction process is performed on each of the connected components in the intermediate mask image to obtain the first centerline of each connected component. The third mask image is determined based on each of the first center lines and the intermediate mask image.

3. The method according to claim 2, characterized in that, Determining the third mask image based on each of the first center lines and the intermediate mask image includes: The centerline extraction process is performed on the tissue to be segmented in the first mask image to obtain the second centerline corresponding to the tissue to be segmented. Calculate the degree of overlap between each of the first centerlines and the second centerlines to obtain multiple degrees of overlap; The third mask image is determined based on the multiple overlaps and the intermediate mask image.

4. The method according to claim 3, characterized in that, Each chamber includes at least a first chamber and a second chamber; determining the third mask image based on the plurality of overlaps and the intermediate mask image includes: The connected region corresponding to the first center line with the highest degree of overlap is determined as the first chamber, and the connected region corresponding to the first center line with the second highest degree of overlap is determined as the second chamber. Based on the determined first and second chambers, each connected region in the intermediate mask image is marked with its respective chamber to obtain the third mask image.

5. The method according to claim 1, characterized in that, The step of determining the first mask image of the tissue to be segmented based on the acquired original image includes: Obtain the original image; The tissue to be segmented in the original image is segmented using a segmentation model or segmentation algorithm to obtain the first mask image, which is a segmented image of the tissue to be segmented.

6. The method according to claim 3, characterized in that, The calculation of the overlap between each of the first centerlines and the second centerlines yields multiple overlap degrees, including: The position information of discrete points on each of the first center lines is compared with the position information of discrete points on each of the second center lines to determine the number of discrete points with the same position information between each of the first center lines and each of the second center lines. The degree of overlap between each of the first center lines and the second center lines is determined based on the number of identical discrete points, thereby obtaining the plurality of degrees of overlap.

7. A chamber division device, characterized in that, The device includes: The first determining module is used to determine a first mask image of the tissue to be segmented based on the acquired original image; wherein the tissue to be segmented includes undivided chambers and intermediate tissues between the chambers, and each chamber includes a true chamber and a false chamber; The second determining module is used to determine an initial mask image of the intermediate tissue based on the original image and the first mask image; the intermediate tissue in the initial mask image includes at least one hole; the hole in the initial mask image is filled to determine a second mask image; The segmentation module is configured to determine first location information of the intermediate tissue based on the second mask image; divide the first mask image into at least two connected components based on the first location information to obtain an intermediate mask image; and determine a third mask image based on each of the divided connected components and the intermediate mask image; wherein the third mask image includes each of the divided cavities.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

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

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

Images