CN115829979A - Breach detection method, apparatus, device, storage medium, and program product - Google Patents
Breach detection method, apparatus, device, storage medium, and program product Download PDFInfo
- Publication number
- CN115829979A CN115829979A CN202211595079.8A CN202211595079A CN115829979A CN 115829979 A CN115829979 A CN 115829979A CN 202211595079 A CN202211595079 A CN 202211595079A CN 115829979 A CN115829979 A CN 115829979A
- Authority
- CN
- China
- Prior art keywords
- intermediate tissue
- breach
- mask image
- determining
- area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Image Processing (AREA)
Abstract
The present application relates to a breach detection method, apparatus, device, storage medium and program product. The method comprises the following steps: determining a first mask image of the intermediate tissue according to the obtained original image; the first mask image comprises the intermediate tissue and an initial breach region on the intermediate tissue; completing the initial crevasse area on the intermediate tissue, and determining a second mask image; the second mask image comprises complete intermediate tissues; processing the second mask image according to the original image to determine at least one candidate breach area; and determining a target break area according to the at least one candidate break area. The method can improve the accuracy of the detection result of the crevasses.
Description
Technical Field
The present application relates to the field of image technologies, and in particular, to a method, an apparatus, a device, a storage medium, and a program product for detecting a breach.
Background
The aortic dissection refers to the process in which blood in the aortic lumen enters the aortic media from the lacerated part of the aortic intimal tear to separate the media, and expands along the major axis direction of the aorta to form a true-false two-lumen separation state of the aortic wall, i.e. the aorta is separated into a true lumen and a false lumen. Where blood can flow between the true and false lumens of the aorta. By detecting a breach in the aortic dissection, an effective data basis can be provided for subsequent further image analysis.
At present, when a crevasse in an aortic dissection is detected, the image values of the true and false cavities can be compared to distinguish the true and false cavities, and the connection part in the middle of the true and false cavities is used as the crevasse to realize the detection of the crevasse.
However, the above-mentioned techniques have a problem that the obtained breach detection result is not accurate enough because the breach situation in the actual scene is complicated.
Disclosure of Invention
In view of the above, it is desirable to provide a breach detection method, apparatus, device, storage medium, and program product capable of improving accuracy of a breach detection result.
In a first aspect, the present application provides a breach detection method, including:
determining a first mask image of the intermediate tissue according to the obtained original image; the first mask image comprises an intermediate tissue and an initial crevasse area on the intermediate tissue;
completing the initial crevasse area on the intermediate tissue, and determining a second mask image; the second mask image comprises a complete intermediate tissue;
processing the second mask image according to the original image to determine at least one candidate breach area;
and determining a target breach zone according to the at least one candidate breach zone.
In one embodiment, the determining the target breach zone according to at least one candidate breach zone includes:
classifying at least one candidate breach area by adopting a preset classification network to determine a target breach area;
the classification network is obtained by training based on a plurality of sample mask images, each sample mask image comprises a sample break area and a standard category of the sample break area, and the standard category is used for representing whether the sample break area is a target break area.
In one embodiment, the completing the initial breach region on the intermediate tissue and determining the second mask image includes:
according to the position information of each point on the intermediate tissue in the first mask image, performing curved surface reconstruction processing on the intermediate tissue and the initial crevasse area to determine a closed curved surface corresponding to the intermediate tissue;
and determining a second mask image according to the closed curved surface and the first mask image.
In one embodiment, the performing curved surface reconstruction processing on the intermediate tissue and the initial crevasse region according to the position information of each point on the intermediate tissue in the first mask image to determine the closed curved surface corresponding to the intermediate tissue includes:
performing point cloud sparse sampling processing on the intermediate tissue and the initial crevasse area according to the position information of each point on the intermediate tissue in the first mask image to determine point cloud data corresponding to the intermediate tissue;
and performing curved surface reconstruction processing on the point cloud data to determine a closed curved surface corresponding to the intermediate tissue.
In one embodiment, the performing curved surface reconstruction processing on the point cloud data to determine the closed curved surface corresponding to the intermediate tissue includes:
calculating the normal vector direction of each point in the point cloud data, and adjusting the normal vector direction of each point to the same direction;
and performing curved surface reconstruction processing on the point cloud data after the normal vector direction is adjusted, and determining a closed curved surface corresponding to the intermediate tissue.
In one embodiment, the processing the second mask image according to the original image to determine at least one candidate breach region includes:
and carrying out image block interception processing on the second mask image according to the original image, and determining at least one candidate breach area.
In one embodiment, the performing image block truncation processing on the second mask image according to the original image to determine at least one candidate breach region includes:
determining the corresponding image value of each point on the original image on the complete intermediate tissue according to the original image;
determining at least one candidate point in each point according to the image value of each point;
and (3) carrying out image block interception processing at the position of at least one candidate point in the second mask image, and determining at least one candidate breach area.
In one embodiment, the determining the first mask image of the intermediate tissue according to the acquired original image includes:
determining a mask image of a tissue to be segmented according to the obtained original image; wherein the tissue to be segmented comprises intermediate tissue;
a first mask image of the intermediate tissue is determined from the original image and the mask image of the tissue to be segmented.
In one embodiment, the intermediate tissue is an intima or a blood vessel.
In a second aspect, the present application further provides a breach detection device, comprising:
the first mask determining module is used for determining a first mask image of the intermediate tissue according to the acquired original image; the first mask image comprises an intermediate tissue and an initial crevasse area on the intermediate tissue;
the second mask determining module is used for completing the initial crevasse area on the intermediate tissue and determining a second mask image; the second mask image comprises a complete intermediate tissue;
the processing module is used for processing the second mask image according to the original image and determining at least one candidate breach area;
and the detection module is used for determining a target breach area according to the at least one candidate breach area.
In a third aspect, the present application further provides a computer device, where the computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program:
determining a first mask image of the intermediate tissue according to the obtained original image; the first mask image comprises an intermediate tissue and an initial crevasse area on the intermediate tissue;
completing the initial crevasse area on the intermediate tissue, and determining a second mask image; the second mask image comprises a complete intermediate tissue;
processing the second mask image according to the original image to determine at least one candidate breach area;
and determining a target breach zone according to the at least one candidate breach zone.
In a fourth aspect, the present application further provides a computer readable storage medium, having a computer program stored thereon, which when executed by a processor, performs the steps of:
determining a first mask image of the intermediate tissue according to the obtained original image; the first mask image comprises an intermediate tissue and an initial crevasse area on the intermediate tissue;
completing the initial crevasse area on the intermediate tissue, and determining a second mask image; the second mask image comprises a complete intermediate tissue;
processing the second mask image according to the original image to determine at least one candidate breach area;
and determining a target breach zone according to the at least one candidate breach zone.
In a fifth aspect, the present application also provides a computer program product, a computer program product comprising a computer program which, when executed by a processor, performs the steps of:
determining a first mask image of the intermediate tissue according to the obtained original image; the first mask image comprises an intermediate tissue and an initial crevasse area on the intermediate tissue;
completing the initial crevasse area on the intermediate tissue, and determining a second mask image; the second mask image comprises a complete intermediate tissue;
processing the second mask image according to the original image to determine at least one candidate breach area;
and determining a target breach zone according to the at least one candidate breach zone.
According to the method, the device, the equipment, the storage medium and the program product for detecting the crevasses, the first mask image of the intermediate tissue is determined through the original image, the initial crevasses on the intermediate tissue are supplemented to determine the second mask image of the complete intermediate tissue, then the second mask image is processed through the original image to obtain at least one candidate crevasses, and the target crevasses are determined according to the candidate crevasses. According to the method, the initial crevasse region of the intermediate tissue can be supplemented, the candidate crevasse region is determined by combining the original image on the mask image of the supplemented intermediate tissue, and the target crevasse region is determined through the candidate crevasse region, so that the problem that the accuracy of the detected crevasse region is low when an unreal crevasse region exists in the initial crevasse region can be avoided, the problem can be avoided by supplementing the intermediate tissue and further detecting the crevasse region, and the accuracy of the finally detected crevasse region is improved.
Drawings
FIG. 1 is an exemplary diagram of a prior art aortic dissection;
FIG. 2 is a diagram of the internal structure of a computer device in one embodiment;
FIG. 3 is a schematic flow chart of a breach detection method in one embodiment;
FIG. 4 is a schematic flow chart of a breach detection method in another embodiment;
FIG. 5 is a schematic flow chart of a breach detection method in another embodiment;
FIG. 6 is an exemplary diagram illustrating sparse sampling of intervening tissue before and after in another embodiment;
FIG. 7 is an exemplary illustration of the mid-tissue normal vector direction adjustment before and after another embodiment;
FIG. 8 is an illustration of an intermediate tissue reconstruction session in another embodiment;
FIG. 9 is a schematic flowchart of another embodiment of a breach detection method;
FIG. 10 is a schematic flow chart of another embodiment of a breach detection method;
fig. 11 is a block diagram showing the structure of a breach detection apparatus according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
After a typical endarterial aortic tear, aortic blood flow can flow into the aortic wall through the ostium to divide the original single-Lumen structure of the aorta into True Lumen (TL) and False Lumen (FL). Referring to fig. 1, the true and false cavities may be communicated or not, the place where the true and false cavities are communicated is a rupture port (i.e. a frame-selecting position in the figure), and a re-rupture port may exist at the far end; when the true and false cavities are not communicated, the two cavities are separated by the aortic intima. At present, when a crevasse in an aortic dissection is detected, the image values of the true and false cavities can be compared to distinguish the true and false cavities, and the connection part in the middle of the true and false cavities is used as the crevasse to realize the detection of the crevasse. However, the above-mentioned techniques have a problem that the obtained breach detection result is not accurate enough because the breach situation in the actual scene is complicated. Based on this, embodiments of the present application provide a breach detection method, apparatus, device, storage medium, and program product, which can solve the above technical problems.
The breach detection method provided by the embodiment of the application can be applied to computer equipment, the computer equipment can be a terminal or a server, taking the terminal as an example, and the internal structure diagram can be shown in fig. 2. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless communication can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement a breach detection method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.
Those skilled in the art will appreciate that the architecture shown in fig. 2 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, as shown in fig. 3, a breach detection method is provided, which is exemplified by the method applied to the computer device in fig. 1, and the method may include the following steps:
s202, determining a first mask image of the intermediate tissue according to the acquired original image; the first mask image includes an intermediate structure and an initial breach region on the intermediate structure.
The original image may be an original image including an intermediate tissue, and the original image of the intermediate tissue is not acquired separately, but is acquired by acquiring an original image of a tissue to be segmented including the intermediate tissue. The tissue to be segmented here may be, for example, the aorta, the carotid artery or the iliac artery, but of course, other tissues in which multiple cavities occur due to dissection and intervening tissues exist. As an alternative embodiment, the intermediate tissue may be an inner membrane or a blood vessel.
The original image may be an image obtained by scanning and reconstructing a tissue to be segmented in real time in advance, or may be an image obtained from an image of a tissue to be segmented pre-stored in a cloud or a server, or may be another manner of obtaining an original image, which is not limited specifically here. The original image may be a CTA (CT angiography) image, or may be another type of CT (Computed Tomography) image or MRI (Magnetic Resonance Imaging) image.
After obtaining the original image of the tissue to be segmented, the tissue to be segmented and the intermediate tissue in the original image may be segmented by using a segmentation model or a segmentation algorithm, etc., to obtain a first mask image of the intermediate tissue. The intermediate tissue included in the first mask image is not a complete intermediate tissue, and there may be some initial crevasses on the intermediate tissue due to segmentation or actual crevasses, and the number of the initial crevasses may be one or more, and the initial crevasses may be a region formed by one or more points.
S204, completing the initial crevasse area on the intermediate tissue, and determining a second mask image; the second mask image includes the entire intermediate structure.
In this step, after obtaining the intermediate tissue including the initial crevasse region, a morphological manner (e.g., a dilation operation or a morphological opening operation), or a mesh filling, or a planar or curved surface reconstruction, etc. may be adopted to perform a hole filling operation on the initial crevasse region on the intermediate tissue, so as to fill the intermediate tissue into a completed intermediate tissue without crevasses.
Specifically, in this step, the initial breach region of the intermediate tissue may be completed on the first mask image to obtain a complete intermediate tissue, that is, to obtain the second mask image.
S206, processing the second mask image according to the original image, and determining at least one candidate breach area.
In this step, the second mask image is obtained after the first mask image obtained from the original image, and then the second mask image and the position information of the intermediate tissue in the original image correspond to each other, and generally, the information about the intermediate tissue portion in the original image is relatively rich, while the mask image is a binarized image and has relatively less information about the intermediate tissue portion. Then the image information of the corresponding position on the original image can be obtained through the relative position of the intermediate tissue, and then the intermediate tissue on the second mask image is subjected to image processing by combining the image information, so as to divide at least one candidate breach area from the intermediate tissue of the second mask image.
As an alternative embodiment, the second mask image may be subjected to image block truncation according to the original image, and at least one candidate breach area is determined, that is, according to image information at a position corresponding to the intermediate tissue on the original image, the intermediate tissue on the second mask image is subjected to image block truncation, and at least one image block is obtained, and the image block may be used as the candidate breach area.
And S208, determining a target break area according to the at least one candidate break area.
In this step, after obtaining at least one candidate breach region, each candidate breach region may be detected, and it is determined whether the candidate breach region is a real breach region or a non-real breach region caused by an error due to dividing the intermediate tissue, and finally, a real breach region may be selected from the candidate breach region through a determination result and marked as a target breach region. The target breach zone may be one or more.
Further, after the target breach area is obtained, the target breach area may be directly pushed to the doctor end for display, or a bounding box image of the bounding box may be captured from the original image and displayed to the doctor end with the target breach area as the center.
In the method for detecting the crevasses, a first mask image of the intermediate tissue is determined through an original image, an initial crevasse area on the intermediate tissue is supplemented to determine a second mask image of the complete intermediate tissue, then the second mask image is processed through the original image to obtain at least one candidate crevasse area, and a target crevasse area is determined according to the candidate crevasse area. According to the method, the initial crevasse region of the intermediate tissue can be supplemented, the candidate crevasse region is determined by combining the original image on the mask image of the supplemented intermediate tissue, and the target crevasse region is determined through the candidate crevasse region, so that the problem that the accuracy of the detected crevasse region is low when an unreal crevasse region exists in the initial crevasse region can be avoided, the problem can be avoided by supplementing the intermediate tissue and further detecting the crevasse region, and the accuracy of the finally detected crevasse region is improved.
In the above embodiments, it is mentioned that the target breach zone can be determined by at least one candidate breach zone, and in order to improve the efficiency and accuracy of determining the target breach zone, the determination may be performed in combination with a classification network, and the following embodiments will describe the process in detail.
In another embodiment, another breach detection method is provided, and on the basis of the foregoing embodiment, the foregoing S208 may include the following steps:
and step A, classifying at least one candidate breach area by adopting a preset classification network, and determining a target breach area.
The classification network may be a neural network, and is not limited to a specific network architecture or a network type. The classification network may be a two-classification network, and is mainly used for distinguishing whether the candidate breach region is a target breach region or a non-target breach region.
Before the classification network is used to classify the candidate breach regions, the classification network may be trained in advance, and for the training process, the classification network is obtained by training based on a plurality of sample mask images, each sample mask image includes a sample breach region and a standard class of the sample breach region, and the standard class is used to characterize whether the sample breach region is a target breach region. The standard type here may be embodied by way of identification, for example, if the standard type of one sample breach region is 1, the sample breach region is regarded as a target breach region, and if the standard type of another sample breach region is 0, the sample breach region is regarded as a non-target breach region; the target breach zone can be considered as a real breach zone.
In the training process of the initial classification network, the prediction category of the sample breach area can be obtained by inputting the sample mask image into the initial classification network, then the loss between the prediction category and the corresponding standard category can be calculated, and the parameter adjustment of the initial classification network is realized through loss feedback. And under the condition that the loss reaches a threshold value or is stable and the like, fixing the parameters of the initial classification network to obtain the trained classification network.
After the classification network is trained, each obtained candidate breach region can be input into the classification network, a classification result corresponding to each candidate breach region is obtained, and the classification result can indicate whether the breach region is a target breach region, so that the target breach region in the candidate breach regions can be determined through the classification result.
In this embodiment, the classification network is used to classify the candidate breach regions to obtain the target breach region, so that the efficiency and accuracy of determining the target breach region can be improved. In addition, the classification network is obtained by training based on a plurality of sample mask images, and each sample mask image comprises a sample break area and a standard category of the sample break area, so that the training of the classification network is performed through a plurality of samples and labels thereof, the trained classification network can be more accurate, and the classification accuracy is further improved.
In the above embodiments, it is mentioned that the initial breach region of the intermediate tissue can be completed, and the following embodiments describe in detail how the completion process is specifically performed.
In another embodiment, another breach detection method is provided, and on the basis of the foregoing embodiment, as shown in fig. 4, the foregoing S204 may include the following steps:
s302, according to the position information of each point on the intermediate tissue in the first mask image, performing curved surface reconstruction processing on the intermediate tissue and the initial breach area, and determining a closed curved surface corresponding to the intermediate tissue.
In this step, when the first mask image is obtained, the position information of each point on the intermediate tissue may also be obtained, and then the initial breach region and the non-initial breach region on the intermediate tissue are subjected to surface reconstruction processing by using any one or more of a surface reconstruction method, an interpolation method, a least square method and the like according to the position information of each point, so that the intermediate tissue is completely reconstructed into a complete closed surface. The closed curved surface has no crevasse area.
S304, determining a second mask image according to the closed curved surface and the first mask image.
In this step, the curved surface reconstruction process may be performed on the intermediate tissue on the first mask image, so that after the closed curved surface of the intermediate tissue is obtained, a mask image including the closed curved surface, that is, a second mask image, is obtained.
In this embodiment, the curved surface reconstruction is performed on the initial crevasse region through the position information of each point on the intermediate tissue in the first mask image, and the second mask image is obtained after the closed curved surface of the intermediate tissue is obtained.
For the specific implementation of the curved surface reconstruction process mentioned in the above embodiments to obtain the closed curved surface of the intermediate tissue, the following embodiments provide one possible implementation.
In another embodiment, another breach detection method is provided, and on the basis of the above embodiment, as shown in fig. 5, the step S302 may include the following steps:
s402, according to the position information of each point on the intermediate tissue in the first mask image, performing point cloud sparse sampling processing on the intermediate tissue and the initial breach area, and determining point cloud data corresponding to the intermediate tissue.
As mentioned above in S302, when obtaining the first mask image, the intermediate tissue in the first mask image can be regarded as a set of many voxels, and then the position information of each voxel therein can be obtained here, where each voxel on the intermediate tissue is relatively dense. Therefore, in order to reduce the calculation amount of subsequent curved surface reconstruction, point cloud sparse sampling processing may be performed on the intermediate tissue and the initial breach area in the first mask image, specifically, each voxel point of the intermediate tissue may be uniformly and sparsely sampled, and each sampled voxel point is obtained, where data formed by each sampled voxel point may be recorded as point cloud data of the intermediate tissue.
By taking the example that the intermediate tissue is an inner membrane as an example, referring to fig. 6, where the left image is a first mask image of the intermediate tissue, the intermediate tissue is an original image before sparse sampling, and the right image is an intermediate tissue after sparse sampling, it can be seen that the voxel points of the intermediate tissue after sparse sampling are distributed more uniformly and in a smaller number, so that the calculation amount for subsequently reconstructing the intermediate tissue is less, and therefore the reconstruction efficiency can be improved.
S404, performing curved surface reconstruction processing on the point cloud data, and determining a closed curved surface corresponding to the intermediate tissue.
After point cloud data obtained by sparsely sampling the intermediate tissue is obtained, the pointing direction of each point in the point cloud data may be inconsistent, so that curved surface reconstruction is not convenient to realize, and reconstruction accuracy is affected. Therefore, in order to avoid this problem, as an alternative embodiment, the normal vector direction of each point in the point cloud data may be calculated, and the normal vector directions of each point may be adjusted to the same direction; and performing curved surface reconstruction processing on the point cloud data after the normal vector direction is adjusted, and determining a closed curved surface corresponding to the intermediate tissue.
That is, after point cloud data obtained by sparsely sampling the intermediate tissue is obtained, a surface centered on the current point may be fitted by the positions of each point and its neighboring points, and the curvature or curvature of each point may be obtained at the same time, and then the normal vector to each point and its direction may be calculated by the least square method or the like. Then, the normal vector direction of each point can be corrected by the normal vector of the point close to each point, so that the normal vectors of all the points point to the same plane, namely point to the same direction, for example, point to the front of a curved surface.
For example, taking the middle tissue as an inner membrane as an example, as shown in fig. 7, a left image is an image in which normal vectors of each point on the middle tissue point to different directions, and a right image is an image in which the normal vectors of each point on the middle tissue point to the same plane in the same direction after the direction is adjusted.
After the normal vectors of all points in the point cloud data are adjusted to the same plane, the point cloud data can be subjected to curved surface reconstruction processing by adopting a reconstruction algorithm, and the intermediate tissue and the initial crevasse area on the intermediate tissue are all reconstructed into a complete closed curved surface to obtain a closed curved surface corresponding to the intermediate tissue. Optionally, the reconstruction algorithm may be a poisson reconstruction algorithm, and the point cloud sparse processing and the poisson reconstruction algorithm are used for performing curved surface reconstruction processing, so that a smooth filling curved surface corresponding to the intermediate tissue can be obtained, and the judgment of the subsequent breach area of the intermediate tissue is more reasonable.
Exemplarily, taking the intermediate tissue as an inner membrane as an example, as shown in fig. 8, a left graph is a closed curved surface of the reconstructed intermediate tissue, wherein the closed curved surface includes a portion where the intermediate tissue after the curved surface reconstruction coincides with the intermediate tissue obtained by segmentation in the first mask image, and the portion belongs to the complete intermediate tissue; the middle image is the left end surface of the reconstructed middle tissue, and the reconstructed middle tissue is in a dotted frame; the right graph is the left end face of the middle tissue before reconstruction, and the middle tissue before reconstruction is in a dotted line frame; it can be seen that the reconstructed intermediate tissue is more complete and is also a closed curved surface without crevasses. In addition, as can be seen from fig. 8, the aorta can be divided into one or more connected domains by the intima-media sheet, which can be used for true and false lumen segmentation of the aorta, and meanwhile, pre-screening of the lacerated area of the intima-media sheet can be realized.
In the embodiment, the point cloud sparse sampling processing is firstly carried out on the initial crevasse area through the position of each point on the intermediate tissue in the first mask image, the point cloud data of the intermediate tissue is determined, the curved surface reconstruction processing is carried out on the point cloud data to obtain the corresponding closed curved surface, the voxel points of the intermediate tissue after sparse sampling are distributed more uniformly, the quantity is less, the subsequent calculation amount for reconstructing the intermediate tissue is less, and therefore the reconstruction efficiency can be improved. In addition, the normal vector directions of all points in the point cloud data are adjusted to the same direction and then curved surface reconstruction processing is carried out, so that a closed curved surface of an intermediate tissue is obtained, the point cloud data after the direction is adjusted are uniform, and subsequent curved surface reconstruction processing can be carried out more conveniently.
For the specific implementation of determining the candidate breach region on the mask image for completing the intermediate tissue mentioned in the above embodiment, the following embodiment provides one possible implementation.
In another embodiment, another breach detection method is provided, and on the basis of the above embodiment, as shown in fig. 9, the step S206 may include the following steps:
s502, according to the original image, determining the corresponding image value of each point on the original image on the complete intermediate tissue.
The original image generally includes image values (e.g., CT values, HU values, etc.), and the second mask image after the hole filling of the intermediate tissue is generally a binarized mask image, on which the image values such as CT values, HU values, etc. are generally not included.
As mentioned above, the second mask image corresponds to the positions of the respective points on the intermediate tissue in the original image, and then the image values of the respective points on the intermediate tissue can be obtained on the original image by corresponding the positions of the respective points on the second mask image to the original image.
And S504, determining at least one candidate point in each point according to the image value of each point.
In this step, a threshold value may be obtained from the image values of the respective points on the original image, and for example, the average HU value of the blood vessel may be used as the threshold value.
Then, the image values of the points on the intermediate tissue may be compared with the threshold values, respectively, to obtain the comparison result of each point. If the image value of one point is less than or equal to the threshold, the point may be considered as an intermediate structure, and if the image value of another point is greater than the threshold, the point may be considered as a point suspected to be a break, and the point suspected to be a break may be considered as a candidate point.
S506, image block interception processing is carried out at the position of at least one candidate point in the second mask image, and at least one candidate breach area is determined.
After the candidate points suspected of being breached are obtained, the positions of the candidate points can be corresponded to the second mask image of the complete intermediate tissue to obtain the corresponding positions of the points; then, image block interception processing can be performed at the position of the corresponding point on the second mask image according to a certain image block size, so as to obtain an image block corresponding to the candidate point on the second mask image, and the image block can be marked as a candidate breach area.
In this embodiment, by obtaining the image values of the points on the original image of the complete intermediate tissue, and determining the candidate points from the points, the image block of the candidate points sitting in the area can be captured from the second mask image, and the candidate breach area can be obtained, so that the suspected breach area on the complete intermediate tissue can be quickly obtained through the image values, and the speed of detecting the breach area can be increased.
In the above embodiments, it is mentioned that a mask image of the intermediate tissue including the initial breach region can be obtained from the original image, and the following embodiments describe in detail how to determine the mask image.
In another embodiment, another breach detection method is provided, and on the basis of the foregoing embodiment, as shown in fig. 10, the foregoing S202 may include the following steps:
s602, determining a mask image of a tissue to be segmented according to the acquired original image; wherein, the tissue to be segmented comprises intermediate tissue.
S604, determining a first mask image of the intermediate tissue according to the original image and the mask image of the tissue to be segmented.
As mentioned in S202 above, after the original image of the tissue to be segmented is obtained, the tissue to be segmented in the original image may be subjected to segmentation processing using a segmentation model or a segmentation algorithm, etc., to obtain a segmented image of the tissue to be segmented, which may be a binarized mask image including the tissue to be segmented and a background. By means of the binary mask image of the tissue to be segmented, background information except the tissue to be segmented can be eliminated, and accuracy of determining the intermediate tissue below is improved. In addition, the segmentation model for segmenting the tissue to be segmented may be a neural network model, for example, a V-net network, and the specific model type is not limited herein. In the training process of the segmentation model of the tissue to be segmented, samples including intermediate tissues and samples not including the intermediate tissues can be input, and the Dice loss is adopted for training.
Further, since the intermediate tissue may be a part of the tissue to be segmented, the original image and the segmented image of the tissue to be segmented may be subjected to a segmentation process of the intermediate tissue by using a segmentation model or a segmentation algorithm, so as to obtain a first mask image of the intermediate tissue.
In addition, the intermediate tissue generally appears in the lumen in a form (e.g., a shaded area/a darker area) different from the lumen in the tissue to be segmented, and the characteristics are significant, so that a superior segmentation effect on the intermediate tissue can be achieved by using the original image and the segmented image of the tissue to be segmented. Moreover, the segmentation of the intermediate tissue is carried out by adding the segmentation image of the tissue to be segmented, so that the segmentation model or the segmentation algorithm only focuses on the segmentation processing of the intermediate tissue on the tissue part to be segmented in the original image, and the efficiency and the accuracy of the segmentation are higher. The segmentation model for segmenting the intermediate tissue may be a neural network model, for example, a V-net network, and the specific model type is not limited here. And (3) training the segmentation model of the intermediate tissue by adopting Mask Dice loss in the training process.
In the embodiment, the mask image of the intermediate tissue is determined through the mask image and the original image of the tissue to be segmented, so that the efficiency and the accuracy of the obtained intermediate tissue can be improved.
A specific example is given below to illustrate the technical solution of the embodiment of the present application, and on the basis of the above embodiment, the method may include the following steps:
s1, segmenting an obtained original image to obtain a mask image of a tissue to be segmented; wherein the tissue to be segmented comprises intermediate tissue; the intermediate tissue is an inner membrane or a blood vessel;
s2, segmenting the intermediate tissue according to the original image and the mask image of the tissue to be segmented to obtain a first mask image of the intermediate tissue; the first mask image comprises an intermediate tissue and an initial crevasse area on the intermediate tissue;
s3, performing point cloud sparse sampling processing on the intermediate tissue and the initial crevasse area according to the position information of each point on the intermediate tissue in the first mask image, and determining point cloud data corresponding to the intermediate tissue;
s4, calculating the normal vector direction of each point in the point cloud data, and adjusting the normal vector direction of each point to the same direction;
s5, performing curved surface reconstruction processing on the point cloud data after the normal vector direction is adjusted, and determining a closed curved surface corresponding to the intermediate tissue;
s6, determining a second mask image according to the closed curved surface and the first mask image;
s7, determining corresponding image values of all points on the complete intermediate tissue on the original image according to the original image;
s8, determining at least one candidate point in each point according to the image value of each point;
s9, carrying out image block interception processing at the position of at least one candidate point in the second mask image, and determining at least one candidate breach area;
s10, classifying at least one candidate breach area by adopting a preset classification network, and determining a target breach area; the classification network is obtained by training based on a plurality of sample mask images, each sample mask image comprises a sample break area and a standard category of the sample break area, and the standard category is used for representing whether the sample break area is a target break area.
It should be understood that, although the steps in the flowcharts related to the embodiments described above are shown in sequence as indicated by the arrows, the steps are not necessarily performed in sequence as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a part of the steps in the flowcharts related to the embodiments described above may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the execution order of the steps or stages is not necessarily sequential, but may be rotated or alternated with other steps or at least a part of the steps or stages in other steps.
Based on the same inventive concept, the embodiment of the application also provides a breach detection device for realizing the above related breach detection method. The solution to the problem provided by the apparatus is similar to the solution described in the above method, so the specific limitations in one or more embodiments of the breach detection apparatus provided below can be referred to the limitations of the breach detection method in the above, and are not described herein again.
In one embodiment, as shown in fig. 11, there is provided a breach detection apparatus, comprising: the device comprises a first mask determining module, a second mask determining module, a processing module and a detecting module, wherein:
the first mask determining module is used for determining a first mask image of the intermediate tissue according to the acquired original image; the first mask image comprises an intermediate tissue and an initial crevasse area on the intermediate tissue;
the second mask determining module is used for completing the initial crevasse area on the intermediate tissue and determining a second mask image; the second mask image comprises a complete intermediate tissue;
the processing module is used for processing the second mask image according to the original image and determining at least one candidate breach area;
and the detection module is used for determining a target breach area according to the at least one candidate breach area.
Optionally, the intermediate tissue is an inner membrane or a blood vessel.
In another embodiment, another breach detection device is provided, and on the basis of the above embodiment, the detection module may include:
the classification unit is used for classifying at least one candidate breach region by adopting a preset classification network to determine a target breach region; the classification network is obtained by training based on a plurality of sample mask images, each sample mask image comprises a sample break area and a standard category of the sample break area, and the standard category is used for representing whether the sample break area is a target break area.
In another embodiment, another apparatus for detecting a breach is provided, and on the basis of the foregoing embodiment, the second mask determining module may include:
the reconstruction unit is used for carrying out curved surface reconstruction processing on the intermediate tissue and the initial crevasse area according to the position information of each point on the intermediate tissue in the first mask image and determining a closed curved surface corresponding to the intermediate tissue;
and the image determining unit is used for determining a second mask image according to the closed curved surface and the first mask image.
In another embodiment, another breach detection apparatus is provided, and on the basis of the above embodiment, the reconstruction unit may include:
the point cloud processing subunit is used for performing point cloud sparse sampling processing on the intermediate tissue and the initial crevasse area according to the position information of each point on the intermediate tissue in the first mask image to determine point cloud data corresponding to the intermediate tissue;
and the reconstruction subunit is used for performing curved surface reconstruction processing on the point cloud data and determining a closed curved surface corresponding to the intermediate tissue.
Optionally, the reconstruction subunit is specifically configured to calculate a normal vector direction of each point in the point cloud data, and adjust the normal vector directions of the points to the same direction; and performing curved surface reconstruction processing on the point cloud data after the normal vector direction is adjusted, and determining a closed curved surface corresponding to the intermediate tissue.
In another embodiment, on the basis of the foregoing embodiment, the processing module is specifically configured to perform image block truncation processing on the second mask image according to the original image, and determine at least one candidate breach area.
Optionally, the processing module may include
The image value determining unit is used for determining the corresponding image value of each point on the complete intermediate tissue on the original image according to the original image;
a candidate point determining unit for determining at least one candidate point of each point according to the image value of each point;
and the intercepting unit is used for carrying out image block intercepting processing at the position of at least one candidate point in the second mask image and determining at least one candidate breach area.
In another embodiment, another apparatus for detecting a breach is provided, and on the basis of the foregoing embodiment, the first mask determining module may include:
the tissue mask determining unit is used for determining a mask image of the tissue to be segmented according to the acquired original image; wherein the tissue to be segmented comprises intermediate tissue;
and the first mask determining unit is used for determining a first mask image of the intermediate tissue according to the original image and the mask image of the tissue to be segmented.
All or part of the modules in the breach detection device can be realized by software, hardware and combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:
determining a first mask image of the intermediate tissue according to the obtained original image; the first mask image comprises an intermediate tissue and an initial crevasse area on the intermediate tissue; completing the initial crevasse area on the intermediate tissue, and determining a second mask image; the second mask image comprises a complete intermediate tissue; processing the second mask image according to the original image to determine at least one candidate breach area; and determining a target breach zone according to the at least one candidate breach zone.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
classifying at least one candidate breach area by adopting a preset classification network to determine a target breach area; the classification network is obtained by training based on a plurality of sample mask images, each sample mask image comprises a sample break area and a standard category of the sample break area, and the standard category is used for representing whether the sample break area is a target break area.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
according to the position information of each point on the intermediate tissue in the first mask image, performing curved surface reconstruction processing on the intermediate tissue and the initial crevasse area to determine a closed curved surface corresponding to the intermediate tissue; and determining a second mask image according to the closed curved surface and the first mask image.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
performing point cloud sparse sampling processing on the intermediate tissue and the initial breach area according to the position information of each point on the intermediate tissue in the first mask image, and determining point cloud data corresponding to the intermediate tissue; and performing curved surface reconstruction processing on the point cloud data to determine a closed curved surface corresponding to the intermediate tissue.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
calculating the normal vector direction of each point in the point cloud data, and adjusting the normal vector direction of each point to the same direction; and performing curved surface reconstruction processing on the point cloud data after the normal vector direction is adjusted, and determining a closed curved surface corresponding to the intermediate tissue.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
and carrying out image block interception processing on the second mask image according to the original image, and determining at least one candidate crevasse area.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
determining the corresponding image value of each point on the original image on the complete intermediate tissue according to the original image; determining at least one candidate point in each point according to the image value of each point; and carrying out image block interception processing at the position of at least one candidate point in the second mask image to determine at least one candidate breach area.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
determining a mask image of a tissue to be segmented according to the obtained original image; wherein the tissue to be segmented comprises intermediate tissue; a first mask image of the intermediate tissue is determined from the original image and the mask image of the tissue to be segmented.
In one embodiment, the intermediate tissue is an intima or blood vessel.
In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:
determining a first mask image of the intermediate tissue according to the obtained original image; the first mask image comprises an intermediate tissue and an initial crevasse area on the intermediate tissue; completing the initial crevasse area on the intermediate tissue, and determining a second mask image; the second mask image comprises a complete intermediate tissue; processing the second mask image according to the original image to determine at least one candidate breach area; and determining a target breach zone according to the at least one candidate breach zone.
In one embodiment, the computer program when executed by the processor further performs the steps of:
classifying at least one candidate breach area by adopting a preset classification network to determine a target breach area; the classification network is obtained by training based on a plurality of sample mask images, each sample mask image comprises a sample break area and a standard category of the sample break area, and the standard category is used for representing whether the sample break area is a target break area.
In one embodiment, the computer program when executed by the processor further performs the steps of:
according to the position information of each point on the intermediate tissue in the first mask image, performing curved surface reconstruction processing on the intermediate tissue and the initial crevasse area to determine a closed curved surface corresponding to the intermediate tissue; and determining a second mask image according to the closed curved surface and the first mask image.
In one embodiment, the computer program when executed by the processor further performs the steps of:
performing point cloud sparse sampling processing on the intermediate tissue and the initial breach area according to the position information of each point on the intermediate tissue in the first mask image, and determining point cloud data corresponding to the intermediate tissue; and performing curved surface reconstruction processing on the point cloud data to determine a closed curved surface corresponding to the intermediate tissue.
In one embodiment, the computer program when executed by the processor further performs the steps of:
calculating the normal vector direction of each point in the point cloud data, and adjusting the normal vector direction of each point to the same direction; and performing curved surface reconstruction processing on the point cloud data after the normal vector direction is adjusted, and determining a closed curved surface corresponding to the intermediate tissue.
In one embodiment, the computer program when executed by the processor further performs the steps of:
and carrying out image block interception processing on the second mask image according to the original image, and determining at least one candidate breach area.
In one embodiment, the computer program when executed by the processor further performs the steps of:
determining the corresponding image value of each point on the original image on the complete intermediate tissue according to the original image; determining at least one candidate point in each point according to the image value of each point; and carrying out image block interception processing at the position of at least one candidate point in the second mask image to determine at least one candidate breach area.
In one embodiment, the computer program when executed by the processor further performs the steps of:
determining a mask image of a tissue to be segmented according to the obtained original image; wherein the tissue to be segmented comprises intermediate tissue; a first mask image of the intermediate tissue is determined from the original image and the mask image of the tissue to be segmented.
In one embodiment, the intermediate tissue is an intima or blood vessel.
In one embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, performs the steps of:
determining a first mask image of the intermediate tissue according to the obtained original image; the first mask image comprises an intermediate tissue and an initial crevasse area on the intermediate tissue; completing the initial crevasse area on the intermediate tissue, and determining a second mask image; the second mask image comprises a complete intermediate tissue; processing the second mask image according to the original image to determine at least one candidate breach area; and determining a target breach zone according to the at least one candidate breach zone.
In one embodiment, the computer program when executed by the processor further performs the steps of:
classifying at least one candidate breach area by adopting a preset classification network to determine a target breach area; the classification network is obtained by training based on a plurality of sample mask images, each sample mask image comprises a sample break area and a standard category of the sample break area, and the standard category is used for representing whether the sample break area is a target break area.
In one embodiment, the computer program when executed by the processor further performs the steps of:
according to the position information of each point on the intermediate tissue in the first mask image, performing curved surface reconstruction processing on the intermediate tissue and the initial crevasse area to determine a closed curved surface corresponding to the intermediate tissue; and determining a second mask image according to the closed curved surface and the first mask image.
In one embodiment, the computer program when executed by the processor further performs the steps of:
performing point cloud sparse sampling processing on the intermediate tissue and the initial breach area according to the position information of each point on the intermediate tissue in the first mask image, and determining point cloud data corresponding to the intermediate tissue; and performing curved surface reconstruction processing on the point cloud data to determine a closed curved surface corresponding to the intermediate tissue.
In one embodiment, the computer program when executed by the processor further performs the steps of:
calculating the normal vector direction of each point in the point cloud data, and adjusting the normal vector direction of each point to the same direction; and performing curved surface reconstruction processing on the point cloud data after the normal vector direction is adjusted, and determining a closed curved surface corresponding to the intermediate tissue.
In one embodiment, the computer program when executed by the processor further performs the steps of:
and carrying out image block interception processing on the second mask image according to the original image, and determining at least one candidate breach area.
In one embodiment, the computer program when executed by the processor further performs the steps of:
determining the corresponding image value of each point on the original image on the complete intermediate tissue according to the original image; determining at least one candidate point in each point according to the image value of each point; and carrying out image block interception processing at the position of at least one candidate point in the second mask image to determine at least one candidate breach area.
In one embodiment, the computer program when executed by the processor further performs the steps of:
determining a mask image of a tissue to be segmented according to the obtained original image; wherein the tissue to be segmented comprises intermediate tissue; a first mask image of the intermediate tissue is determined from the original image and the mask image of the tissue to be segmented.
In one embodiment, the intermediate tissue is an intima or blood vessel.
It should be noted that the data referred to in the present application (including but not limited to data for analysis, stored data, presented data, etc.) are data that are fully authorized by the respective parties.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high-density embedded nonvolatile Memory, resistive Random Access Memory (ReRAM), magnetic Random Access Memory (MRAM), ferroelectric Random Access Memory (FRAM), phase Change Memory (PCM), graphene Memory, and the like. Volatile Memory can include Random Access Memory (RAM), external cache Memory, and the like. 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), among others. The databases referred to in various embodiments provided herein may include at least one of relational and non-relational databases. The non-relational database may include, but is not limited to, a block chain based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic devices, quantum computing based data processing logic devices, etc., without limitation.
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.
Claims (10)
1. A breach detection method, comprising:
determining a first mask image of the intermediate tissue according to the obtained original image; the first mask image comprises the intermediate tissue and an initial breach region on the intermediate tissue;
completing the initial crevasse area on the intermediate tissue, and determining a second mask image; the second mask image comprises complete intermediate tissues;
processing the second mask image according to the original image to determine at least one candidate breach area;
and determining a target break area according to the at least one candidate break area.
2. The method of claim 1, wherein determining a target breach zone based on the at least one candidate breach zone comprises:
classifying the at least one candidate breach area by adopting a preset classification network to determine a target breach area;
the classification network is obtained by training based on a plurality of sample mask images, each sample mask image comprises a sample break area and a standard category of the sample break area, and the standard category is used for representing whether the sample break area is a target break area.
3. The method of claim 1, wherein said completing the initial breach region on the intermediate tissue and determining a second mask image comprises:
according to the position information of each point on the intermediate tissue in the first mask image, performing curved surface reconstruction processing on the intermediate tissue and the initial crevasse area, and determining a closed curved surface corresponding to the intermediate tissue;
and determining the second mask image according to the closed curved surface and the first mask image.
4. The method according to claim 3, wherein the performing surface reconstruction processing on the intermediate tissue and the initial breach region according to the position information of each point on the intermediate tissue in the first mask image to determine the closed curved surface corresponding to the intermediate tissue comprises:
according to the position information of each point on the intermediate tissue in the first mask image, performing point cloud sparse sampling processing on the intermediate tissue and the initial crevasse area to determine point cloud data corresponding to the intermediate tissue;
and performing curved surface reconstruction processing on the point cloud data, and determining a closed curved surface corresponding to the intermediate tissue.
5. The method of claim 4, wherein performing surface reconstruction processing on the point cloud data to determine a closed surface corresponding to the intermediate tissue comprises:
calculating the normal vector direction of each point in the point cloud data, and adjusting the normal vector direction of each point to the same direction;
and performing curved surface reconstruction processing on the point cloud data after the normal vector direction is adjusted, and determining a closed curved surface corresponding to the intermediate tissue.
6. The method of any of claims 1-5, wherein the processing the second mask image from the raw image to determine at least one candidate breach region comprises:
and carrying out image block interception processing on the second mask image according to the original image to determine at least one candidate breach area.
7. A breach detection device, comprising:
the first mask determining module is used for determining a first mask image of the intermediate tissue according to the acquired original image; the first mask image comprises the intermediate tissue and an initial breach region on the intermediate tissue;
the second mask determining module is used for completing the initial crevasse area on the intermediate tissue and determining a second mask image; the second mask image comprises complete intermediate tissues;
the processing module is used for processing the second mask image according to the original image and determining at least one candidate breach area;
and the detection module is used for determining a target breach area according to the at least one candidate breach area.
8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 6.
9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out 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 realizes the steps of the method of any one of claims 1 to 6 when executed by a processor.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211595079.8A CN115829979A (en) | 2022-12-13 | 2022-12-13 | Breach detection method, apparatus, device, storage medium, and program product |
| PCT/CN2023/138517 WO2024125567A1 (en) | 2022-12-13 | 2023-12-13 | Systems and methods for image segmentation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211595079.8A CN115829979A (en) | 2022-12-13 | 2022-12-13 | Breach detection method, apparatus, device, storage medium, and program product |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115829979A true CN115829979A (en) | 2023-03-21 |
Family
ID=85546732
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211595079.8A Pending CN115829979A (en) | 2022-12-13 | 2022-12-13 | Breach detection method, apparatus, device, storage medium, and program product |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115829979A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2024125567A1 (en) * | 2022-12-13 | 2024-06-20 | Shanghai United Imaging Intelligence Co., Ltd. | Systems and methods for image segmentation |
-
2022
- 2022-12-13 CN CN202211595079.8A patent/CN115829979A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2024125567A1 (en) * | 2022-12-13 | 2024-06-20 | Shanghai United Imaging Intelligence Co., Ltd. | Systems and methods for image segmentation |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110148192B (en) | Medical image imaging method, device, computer equipment and storage medium | |
| CN111161275B (en) | Method and device for segmenting target object in medical image and electronic equipment | |
| WO2021203795A1 (en) | Pancreas ct automatic segmentation method based on saliency dense connection expansion convolutional network | |
| CN111369525B (en) | Image analysis method, apparatus and storage medium | |
| CN108171709A (en) | Detection method, device and the realization device of Liver masses focal area | |
| CN115965750B (en) | Vascular reconstruction method, vascular reconstruction device, vascular reconstruction computer device, and vascular reconstruction program | |
| CN110751187A (en) | Training method of abnormal area image generation network and related product | |
| CN116188485A (en) | Image processing method, device, computer equipment and storage medium | |
| CN115829979A (en) | Breach detection method, apparatus, device, storage medium, and program product | |
| CN113487536A (en) | Image segmentation method, computer device and storage medium | |
| Wen et al. | Pixel-wise triplet learning for enhancing boundary discrimination in medical image segmentation | |
| CN116091466A (en) | Image analysis method, computer device, and storage medium | |
| CN113192031B (en) | Vascular analysis method, vascular analysis device, vascular analysis computer device, and vascular analysis storage medium | |
| CN114170440A (en) | Method and device for determining image feature points, computer equipment and storage medium | |
| CN114240949A (en) | Cervical cell segmentation network training method, cervical cell segmentation method and cervical cell segmentation device | |
| CN115953426A (en) | Cavity segmentation method, device, equipment, storage medium and program product | |
| CN116128895A (en) | Medical image segmentation method, apparatus and computer readable storage medium | |
| CN115760813A (en) | Screw channel generation method, device, equipment, medium and program product | |
| CN116029986A (en) | Method, apparatus, device, storage medium and program product for detecting target tissue | |
| CN114723723A (en) | Medical image processing method, computer device and storage medium | |
| CN114299010A (en) | Method and device for segmenting brain tumor image, computer equipment and storage medium | |
| CN115546089A (en) | Medical image segmentation method, pathological image processing method, device and equipment | |
| CN117115187B (en) | Carotid artery wall segmentation method, carotid artery wall segmentation device, carotid artery wall segmentation computer device, and carotid artery wall segmentation storage medium | |
| CN115661082A (en) | Coronary vessel segmentation method, apparatus, device, storage medium and program product | |
| CN116309438A (en) | Tissue segmentation method, computer device, and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |