Cardiac CT image processing method and device, non-transient computer readable storage medium

Abstract

The present application provides a method and device for processing cardiac CT images, and a non-transitory computer-readable storage medium. The method for processing cardiac CT images includes reconstructing a cardiac CT image of a detected object based on projection data of a cardiac region of the detected object; processing voxel points in the cardiac CT image whose voxel values ​​are within a first threshold range to obtain a first image, wherein the first threshold range is defined by the CT value of a metal object; performing lung region segmentation on the first image to obtain a second image; performing initial cardiac region segmentation on the second image to obtain a third image; and performing rib removal on the third image to obtain a cardiac image.

Description

Technical Field

[0001] The present invention relates to image processing, and in particular to a method and device for processing cardiac CT images, and a non-transitory computer-readable storage medium. Background Art

[0002] Heart disease, especially coronary heart disease, is one of the major diseases that seriously affect human health. Early detection of atherosclerotic plaques, aneurysms and abnormal structures of coronary arteries is an important factor in the diagnosis and treatment of coronary heart disease. In the detection of coronary heart disease, it mainly depends on the three-dimensional image of the heart of the detected object (e.g., the patient). Therefore, how to accurately obtain the three-dimensional image of the heart of the detected object becomes the key.

[0003] The heart is located in the chest cavity and between the two lungs. In the process of obtaining a three-dimensional image of the heart, it is necessary to remove the adjacent tissues around the heart, such as the lungs and ribs. In the process of removing the adjacent tissues, if there are metal objects in the heart, such as a pacemaker, or if the subject has undergone open-chest surgery such as a heart bypass, when the CT image of the subject's heart is segmented, the metal objects and the artifacts they cause will have a serious impact on the segmentation, making the obtained three-dimensional image of the heart inaccurate, for example, including part of the ribs or part of the lung area, or the segmented heart area is incomplete, etc., which will affect the subsequent diagnosis of the coronary artery. Summary of the invention

[0004] The present invention provides a method and device for processing cardiac CT images, and a non-transitory computer-readable storage medium.

[0005] An exemplary embodiment of the present invention provides a method for processing a cardiac CT image, which includes reconstructing a cardiac CT image of an object to be detected based on projection data of a cardiac region of the object to be detected; processing voxel points in the cardiac CT image whose voxel values ​​are within a first threshold range to obtain a first image, wherein the first threshold range is limited by the CT value of a metal object; performing lung region segmentation on the first image to obtain a second image; performing initial cardiac region segmentation on the second image to obtain a third image; and removing ribs from the third image to obtain a cardiac image.

[0006] An exemplary embodiment of the present invention further provides a non-transitory computer-readable storage medium for storing a computer program, wherein the computer program includes instructions for executing the above-mentioned method for processing cardiac CT images.

[0007] An exemplary embodiment of the present invention also provides a processing device for cardiac CT images, which includes an image reconstruction module, a preprocessing module, a lung segmentation module and an initial segmentation module. The image reconstruction module is configured to reconstruct the cardiac CT image of the detected object based on projection data of the cardiac region of the detected object. The preprocessing module is configured to process voxel points in the cardiac CT image whose voxel values ​​are within a first threshold range and obtain a first image, wherein the first threshold range is defined by the CT value of the metal object. The lung segmentation module is configured to perform lung region segmentation on the first image to obtain a second image. The initial segmentation module is configured to perform initial cardiac region segmentation on the second image to obtain a third image. The rib removal module is configured to perform rib removal on the third image to obtain a cardiac image.

[0008] Other features and aspects will become apparent from the following detailed description, drawings, and claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The present invention may be better understood by describing exemplary embodiments of the present invention in conjunction with the accompanying drawings, in which:

[0010] Figure 1 is a schematic diagram of a CT system according to some embodiments of the present invention;

[0011] Figure 2 is based on Figure 1 A schematic diagram of a control system in the CT system shown;

[0012] Figure 3 is a schematic diagram of a cardiac CT image processing device according to some embodiments of the present invention;

[0013] Figure 4 Image (1) shows an initial cardiac CT image, images (2) to (4) show cardiac images obtained by performing cardiac segmentation without processing metal voxels in the prior art, and images (5) to (7) show cardiac images obtained by performing cardiac segmentation after replacing the voxel values ​​of metal objects with preset voxel values ​​in some embodiments of the present invention;

[0014] Figure 5 Images (1) to (8) in the figure respectively show some intermediate images in the processing process of the cardiac CT image processing device in some embodiments of the present invention;

[0015] Figure 6 Images (1) to (5) show heart images obtained by performing heart segmentation using a rib removal method in the prior art, and images (6) to (10) show heart images corresponding to images (1) to (5) and obtained by performing heart segmentation using a rib removal method in some embodiments of the present invention;

[0016] Figure 7 A flowchart showing a method for processing cardiac CT images according to some embodiments of the present invention is shown;

[0017] Figure 8 Shown according to Figure 7 A flowchart of step 240 in the method shown;

[0018] Fig. 9 Shown according to Figure 7 A flowchart of step 250 of the method shown; and

[0019] Fig.10 Shown according to Figure 7 A flow chart of step 260 of the method is shown. DETAILED DESCRIPTION

[0020] The specific embodiments of the present invention will be described below. It should be noted that in the specific description of these embodiments, in order to provide a concise description, it is impossible for this specification to provide a detailed description of all the features of the actual embodiments. It should be understood that in the actual implementation of any embodiment, just as in the process of any engineering project or design project, in order to achieve the specific goals of the developer and to meet system-related or business-related restrictions, various specific decisions are often made, and this will also change from one embodiment to another. In addition, it can also be understood that although the efforts made in this development process may be complex and lengthy, for ordinary technicians in the field related to the content disclosed by the present invention, some changes such as design, manufacturing or production based on the technical content disclosed in this disclosure are just conventional technical means, and should not be understood as insufficient content of this disclosure.

[0021] Unless otherwise defined, the technical or scientific terms used in the claims and the specification shall have the usual meaning understood by persons with ordinary skills in the technical field to which the invention belongs. The words "first", "second" and similar words used in the patent application specification and the claims of the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "One" or "one" and other similar words do not indicate a quantitative limitation, but indicate the existence of at least one. "Include" or "comprises" and other similar words mean that the elements or objects appearing before "include" or "comprises" include the elements or objects listed after "include" or "comprises" and their equivalent elements, and do not exclude other elements or objects. "Connected" or "connected" and other similar words are not limited to physical or mechanical connections, nor are they limited to direct or indirect connections.

[0022] In this article, the term "CT value" is a unit of measurement for determining the density of a local tissue or organ in the human body, usually called Hounsfield unit (HU). The CT value can represent a pixel value or a voxel value.

[0023] Figure 1 FIG. 1 is a schematic diagram of a CT system 10 according to an embodiment of the present invention. Figure 1 , the CT system 10 includes an X-ray source 14, a detector assembly 18, a carrier bed 46, and a control system (not shown). The X-ray source 14 and the detector assembly 18 are located on the gantry 12. The detector assembly 18 is composed of a plurality of detectors and a data acquisition system (DAS) 32. The plurality of detectors sense the attenuated X-ray intensity passing through the object 22 to be inspected and thereby generate analog electrical signals, and the data acquisition system 32 converts the analog electrical signals into digital signals for subsequent processing. During scanning to acquire X-ray projection data, the gantry 12 and the components mounted thereon (e.g., the X-ray source 14 and the detector assembly 18, etc.) rotate around the center of the gantry opening 48. The carrier bed 46 can move the object 22 to be inspected through the gantry opening 48 in whole or in part.

[0024] Figure 2 Shows Figure 1 The control system diagram of the CT system is shown in FIG. Figure 2 As shown, the control system includes an X-ray controller 28 , a gantry controller 30 and a carrier bed controller 44 , wherein the X-ray controller 28 controls the operation of the X-ray source 14 , the gantry controller 30 controls the operation of the gantry 12 , and the carrier bed controller 44 can control the operation of the carrier bed 46 .

[0025] The control system further includes an image reconstruction module 34 , which can receive signals from the data acquisition system 32 and pre-process the signals to reconstruct a CT image of the detected object.

[0026] The control system also includes a computer 36. On the one hand, the computer 36 can store the CT images generated by the image reconstruction module 34 and / or the projection data sent by the data acquisition system 32, and store them in a large-capacity storage 38, which may include computer RAM, disks, etc., and can be displayed to the operator through an associated display 42. On the other hand, the computer 36 can also be connected to the operator console 40 for communication to receive commands or scanning parameters transmitted by the operator through an associated input device to control the X-ray controller 28, the gantry controller 30 and the carrier bed controller 44 to perform related operations. The associated input device includes a form of operator interface such as a keyboard, a mouse, a voice-activated controller or any other suitable input device.

[0027] In order to better detect and / or diagnose the coronary artery, it is necessary to obtain a three-dimensional image of the heart in the phase with the slowest contraction. When the heart region of the detected object is CT scanned, the heartbeat of the detected object is monitored at the same time to obtain the heart waveform of the detected object. In some embodiments, in order to obtain a three-dimensional image of the heart in the phase with the slowest contraction, the detected object can be CT scanned during the cardiac cycle to collect projection data of its heart region. In the image reconstruction stage, based on the heart waveform, the projection data of the appropriate heart interval is selected to reconstruct the three-dimensional image of the heart. In other embodiments, before scanning the detected object, the phase with the slowest contraction of the detected object can be pre-determined. By monitoring the heart waveform of the detected object, the detected object is scanned in the pre-determined phase to reconstruct the three-dimensional image of the heart based on the collected projection data. In some non-limiting embodiments, there will be a peak in the heart waveform, which is called an R wave. The phase with the slowest contraction is generally a period of time after the R wave. The cardiac CT image processing device and method proposed in the present application can perform cardiac segmentation on the three-dimensional cardiac image obtained through the above embodiments.

[0028] Figure 3 FIG. 1 is a schematic diagram of a cardiac CT image processing apparatus 100 according to some embodiments of the present invention. Figure 3 As shown, the processing device 100 for cardiac CT images includes an image reconstruction module 34, a preprocessing module 110, a lung segmentation module 120, an initial segmentation module 130, and a rib removal module 140. The image reconstruction module 34 is used to reconstruct the cardiac CT image of the object to be detected based on the projection data of the cardiac region of the object to be detected. The preprocessing module 110 is used to process the voxel points in the cardiac CT image whose voxel values ​​are within a first threshold range and obtain a first image, wherein the first threshold range is defined by the CT value of the metal object. In some embodiments, the preprocessing module 110 is used to replace the voxel values ​​within the first threshold range with preset voxel values ​​to avoid the influence of metal objects in the heart on the cardiac segmentation. In some embodiments, the preset voxel value includes the CT value of water or other substances. Preferably, the preset voxel value is the CT value of water, for example, 0HU. In this way, the poor accuracy of heart recognition caused by metal artifacts, such as large offset of the heart center or large missing, can be effectively avoided. As Figure 4 As shown in the figure, image (1) shows the initial heart CT image, images (2) to (4) are heart images obtained by heart segmentation in the prior art without processing the metal voxels, and images (5) to (7) are heart images obtained by heart segmentation after replacing the voxel values ​​of the metal object with the voxel values ​​of water, as described above, according to the present invention. Figure 4It can be seen from images (2) to (4) that the metal artifact in the middle is mistakenly identified as lung tissue, resulting in a large deviation of the determined heart center from the actual one, resulting in an incomplete heart image after segmentation. Figure 4 As can be seen from images (5) to (7), the accuracy of heart segmentation is greatly improved due to the removal of metal artifacts.

[0029] In some embodiments, the preprocessing module 110 processes the cardiac CT image based on a binary method to obtain a first image, such as Figure 5 The lung segmentation module 120 is used to perform lung region segmentation on the first image to obtain a second image. The initial segmentation module 130 is used to perform initial heart region segmentation on the second image to obtain a third image, as shown in FIG. Figure 5 The rib removal module 140 is used to remove the ribs from the third image to obtain a heart image, such as Figure 5 As shown in image (8).

[0030] In some non-limiting embodiments, the cardiac CT image processing device 100 further includes a judgment module 150, which is used to judge whether there is a metal object in the heart of the detected object based on the voxel value in the cardiac CT image. Generally, the CT value of the metal object defines a first threshold range. In order to better judge whether there is a metal object in the heart of the detected object, the first threshold range can be set larger. In some embodiments, the judgment module 150 can judge whether there is a metal object in the heart of the detected object by judging whether a certain proportion of voxel values ​​in the cardiac CT image reconstructed by the image reconstruction module 34 are within the first threshold range. In some embodiments, the judgment module 150 is connected to the preprocessing module 110, that is, no matter whether the judgment module 150 judges whether there is a metal object in the heart of the detected object, the preprocessing module 110 processes the voxel points in the cardiac CT image with voxel values ​​within the first threshold range and obtains the first image. When the judgment module 150 judges that there is no metal object in the detected object, it is equivalent to that there are no voxel points in the cardiac CT image within the first threshold range, and the first image obtained by the preprocessing module 110 is the cardiac CT image.

[0031] In other embodiments, the cardiac CT image processing device 100 may not include the judgment module 150. The cardiac CT image processing device 100 may be configured as follows: Figure 2 The computer 36 in the CT system shown is connected to the operator console 40 for communication. The operator can input the presence of a pacemaker or other metal object in the object being detected through relevant input devices, and the processing device 100 can receive the signal and perform subsequent operations on the cardiac CT image.

[0032] In some embodiments, the lung segmentation module 120 includes a lung recognition module 121 and a first extraction module 122 .

[0033] The lung recognition module 121 is connected to the preprocessing module 110 and is used to recognize the region of the voxel points in the first image obtained by the preprocessing module 110 whose voxel values ​​are within the second threshold range as the lung region.

[0034] (like Figure 5 The image (2) in FIG. 1 shows the identified lung region), and further identifies the convex hull of the lung region. In some embodiments, the second threshold range is defined by the CT value of the lung region. In some embodiments, the lung identification module 121 is used to determine the convex hull of the lung region based on a convex hull algorithm. In other embodiments, the lung identification module 121 may also determine the convex hull of the lung region based on other suitable algorithms. By identifying the convex hull of the lung region, for example, Figure 5 Image in (3).

[0035] The first extraction module 122 is connected to the lung recognition module 121 and is used to extract the convex envelope and its internal area recognized by the lung recognition module 121 to obtain a second image. In some embodiments, the internal area of ​​the convex envelope includes the lung area, the heart area, and the ribs, etc. For example, Figure 5 The image (2) in Figure 2 shows the lung region. Figure 5 The image (4) in FIG. 4 shows the inner region of the lung. By identifying and extracting the lung region, the chest wall (e.g. Figure 5 The problem of inaccurate heart segmentation caused by the tissue located outside the lungs in the middle image (1).

[0036] In some embodiments, the initial segmentation module 130 includes a distance calculation module 131 , a heart identification module 132 and a second extraction module 133 .

[0037] The distance calculation module 131 is connected to the first extraction module 122 and is used to calculate the distance from each voxel point not in the lung area to all voxel points in the lung area based on the second image and obtain the shortest distance corresponding to each voxel point. In some embodiments, the distance calculation module 131 can calculate the Euclidean distance (hereinafter referred to as the Euclidean distance) from each voxel point in the non-lung area to the point closest to the voxel point in the lung area. For example, the distance from any voxel point (x1, y1, z1) in the non-lung area to the voxel point (x′1, ′ The calculation formula of the Euclidean distance D1 of (y′1, z′1) is as follows:

[0038]

[0039] That is, in the second image, all voxel points ((x1, y1, z1)~(x n ,y n ,z n )) corresponds to a shortest Euclidean distance D1~D n .

[0040] The heart identification module 132 is used to identify the area where the voxel points in the non-lung area whose corresponding shortest distance is greater than or equal to a preset threshold are located as the initial heart area. In some embodiments, the preset threshold is determined based on the maximum value of the shortest distances corresponding to the voxel points not in the lung area and a predetermined proportionality coefficient, for example, based on the shortest Euclidean distances D1 to D n The maximum value max(D1~D n ) and proportional coefficients α, β, the preset threshold T is determined according to the following formula:

[0041] T=(1+β)·α·max(D1~D n )

[0042] Among them, α and β are both constants. By selecting appropriate values ​​of α and β, it can be ensured that the identified initial heart area retains the shape of the heart while removing the ribs and spine as much as possible.

[0043] In other embodiments, the preset threshold value may be determined according to a proportionality coefficient. In other embodiments, the preset threshold value may be determined according to other formulas.

[0044] Figure 5 Image (5) in FIG. 4 shows the initial heart region identified and extracted. It can be seen from image (5) that, compared with image (4), the non-heart parts in the lower part of image (4), such as the spine, have been removed.

[0045] The second extraction module 133 is used to extract the convex envelope and the initial heart region to obtain a third image.

[0046] In some embodiments, the rib removal module 140 includes a connection point determination module 141 , a rib identification module 142 , and a third extraction module 143 .

[0047] The connection point determination module 141 is used to obtain the first pair of rib connection points according to the tangent points between the convex envelope and the initial heart region, for example Figure 5In the image (6), two points 51 and 52 indicated by the symbol "*" are found (the symbol "*" is only used for illustration to facilitate description, and may not be marked or may be marked in other forms in practice), and a second pair of rib connection points are found in the third image that are closest to the first pair of rib connection points and whose voxel values ​​are within a third threshold range. In some embodiments, the third threshold range is defined based on the CT value of fat. In some non-limiting embodiments, there is a first rib connection point on each of the left and right sides of the third image, and a second rib connection point that meets the conditions can be found near the first rib connection point on the left, and another second rib connection point that meets the conditions can be found near the first rib connection point on the right. For example, when the voxel value of the first rib connection point is 1000, a point that is closest to the first rib connection point and whose voxel value is not equal to 1000 is found as the second rib connection point.

[0048] The rib identification module 142 is used to determine the shortest path between the second pair of rib connection points. In some embodiments, the shortest path is determined by a rule that the path does not pass through the heart region or the rib structure. In some embodiments, the rib identification module 142 can further determine the shortest path between the first pair of rib connection points. In some embodiments, the rib identification module 142 can determine the shortest path based on the Dijkstra algorithm. In other embodiments, the rib identification module 142 can also determine the connection lines between the rib connection points based on other algorithms. Figure 5 The connecting line 60 in the image (7) shows the shortest path between the second pair of rib connection points obtained in one embodiment.

[0049] The third extraction module 143 is used to remove the convex envelope and the shortest path and the area therebetween in the third image to obtain a heart image. In some embodiments, if the rib recognition module 142 further determines the shortest path between the first pair of rib connection points, the third extraction module 143 may also remove the convex envelope and the shortest path between the first pair of rib connection points and the area therebetween in the third image to obtain a heart image such as Figure 5 The heart image shown in image (8) .

[0050] Figure 6 The images (1) to (5) show the heart images obtained by performing heart segmentation using the rib removal method in the prior art. Figure 6 Images (6) to (10) show heart images corresponding to images (1) to (5) obtained by performing heart segmentation according to the rib removal method in some embodiments of the present invention. By comparison, it can be found that the rib removal module 140 proposed in the present application can remove ribs better and more accurately while retaining the complete heart area.

[0051] Figure 7FIG. 2 is a flowchart of a method 200 for processing cardiac CT images according to some embodiments of the present invention. Figure 7 As shown, the method 200 for processing cardiac CT images includes the following steps.

[0052] In step 210, a cardiac CT image of the subject is reconstructed based on projection data of the cardiac region of the subject.

[0053] Optionally, the cardiac CT image processing method 200 further includes step 220, determining whether there is a metal object in the heart of the detected object based on the voxel value in the cardiac CT image. In some embodiments, the determination of whether there is a metal object in the heart of the detected object is performed by determining whether a certain proportion of voxel values ​​in the cardiac CT image are within a first threshold range. In some embodiments, the first threshold range is defined by the CT value of the metal object.

[0054] In step 230, voxel points in the cardiac CT image whose voxel values ​​are within a first threshold range are processed to obtain a first image, wherein the first threshold range is defined by the CT value of the metal object. In some embodiments, processing the voxel points in the cardiac CT image whose voxel values ​​are within the first threshold range includes replacing the voxel values ​​within the first threshold range with preset voxel values ​​to avoid the influence of the metal object in the heart on the cardiac segmentation. In some embodiments, the preset voxel value includes the CT value of water or other substances, and preferably, the preset voxel value is 0HU. In some embodiments, the cardiac CT image can be processed based on a binary method to obtain the first image.

[0055] In step 240 , lung region segmentation is performed on the first image to obtain a second image.

[0056] In step 250, an initial heart region segmentation is performed on the second image to obtain a third image.

[0057] In step 260, rib removal is performed on the third image to obtain a heart image.

[0058] Figure 8 Shows Figure 7 Flowchart of step 240 in the method shown. Figure 8 As shown, step 240 includes the following steps.

[0059] In step 241, a region of voxel points in the first image whose voxel values ​​are within a second threshold range is identified as a lung region, and a convex envelope of the lung region is identified. In some embodiments, the second threshold range is defined by a CT value of the lung region. In some embodiments, the convex envelope of the lung region is determined by a convex hull algorithm.

[0060] In step 242, the convex envelope and its internal area are extracted to obtain a second image. In some embodiments, the internal area of ​​the convex envelope includes a lung area, a heart area, and ribs, etc. In some embodiments, by extracting the convex envelope and its internal area, it is equivalent to removing the area outside the convex envelope (e.g., the chest wall), which can avoid dividing some voxel points in the chest wall into the initial heart area, resulting in inaccurate heart segmentation results.

[0061] Fig. 9 Shows Figure 7 Flowchart of step 250 in the method shown. Fig. 9 As shown, step 250 includes the following steps.

[0062] In step 251, based on the second image, the distance between each voxel point not in the lung region and all voxel points in the lung region is calculated to obtain the shortest distance corresponding to each voxel point. In some embodiments, the shortest distance D1 can be calculated according to the following formula:

[0063]

[0064] Among them, (x1, y1, z1) is any voxel point in the non-lung area, and (x′1, y′1, z′1) is the voxel point in the lung area that is closest to the voxel point (x1, y1, z1).

[0065] In step 252, the region where the voxel points whose corresponding shortest distances are greater than or equal to a preset threshold are located is identified as the initial heart region. In some embodiments, the preset threshold is the product of the maximum value of the shortest distances corresponding to the voxel points not in the lung region and a predetermined proportionality coefficient.

[0066] In step 253, the convex envelope and the initial heart region are extracted to obtain a third image.

[0067] Fig.10 Shows Figure 7 Flowchart of step 260 in the method shown. Fig.10 As shown, step 260 includes the following steps.

[0068] In step 261, a first pair of rib connection points is obtained according to the tangent points between the convex envelope and the initial heart region.

[0069] In step 262, a second pair of rib connection points that are closest to the first pair of rib connection points and whose voxel values ​​are within a third threshold range are correspondingly found in the third image. In some embodiments, the third threshold range is defined based on the CT value of fat.

[0070] In step 263, the shortest path between the second pair of rib connection points is determined. In some embodiments, the shortest path is determined based on the Dijkstra algorithm. In some embodiments, the shortest path is determined based on the rule that the path does not pass through the heart region or the rib structure. In some embodiments, the shortest path between the first pair of rib connection points can also be determined.

[0071] In step 264, the convex envelope and the shortest path and the area therebetween are removed from the third image to obtain a heart image. In some embodiments, the shortest path between the convex envelope and the first pair of rib connection points and the area therebetween may also be removed.

[0072] The cardiac CT image processing method proposed in the present application can better segment the three-dimensional image of the heart of the subject being detected when there are metal objects in the heart of the subject being detected. Moreover, by improving the rib removal algorithm, the ribs can be removed more accurately while retaining the complete heart structure, thereby providing a strong basis for the detection and / or diagnosis of coronary arteries.

[0073] The present invention may also provide a non-transitory computer-readable storage medium for storing an instruction set and / or a computer program, wherein when the instruction set and / or the computer program is executed by a computer, the computer executes the above-mentioned cardiac CT image processing method. The computer executing the instruction set and / or the computer program may be a computer of a CT system, for example Figure 2 In one embodiment, the instruction set and / or computer program may be compiled into a processor / controller of the computer 36.

[0074] Specifically, the set of instructions and / or computer program, when executed by a computer, causes the computer to:

[0075] reconstructing a cardiac CT image of the subject to be detected based on projection data of a cardiac region of the subject to be detected;

[0076] Processing voxel points in the cardiac CT image whose voxel values ​​are within a first threshold range to obtain a first image;

[0077] Performing lung region segmentation on the first image to obtain a second image;

[0078] performing initial heart region segmentation on the second image to obtain a third image;

[0079] The third image is subjected to rib removal to obtain a heart image.

[0080] The instructions described above may be combined into one instruction for execution, and any instruction may be split into multiple instructions for execution. In addition, the execution order of the instructions described above is not limited to the above order.

[0081] In some embodiments, the set of instructions and / or the computer program further causes the computer to:

[0082] It is determined whether there is a metal object in the heart of the detected object based on the voxel value in the heart CT image.

[0083] In some embodiments, performing lung region segmentation on the first image to obtain the second image may include:

[0084] Identify a region of voxel points in the first image whose voxel values ​​are within a second threshold range as a lung region, and identify a convex envelope of the lung region; and

[0085] The convex envelope and its internal area are extracted to obtain a second image.

[0086] In some embodiments, instruction four includes the following instructions:

[0087] Based on the second image, calculating the distance between each voxel point not in the lung region and all voxel points in the lung region to obtain the shortest distance corresponding to each voxel point;

[0088] Identify the region where the corresponding voxel points whose shortest distance is greater than or equal to a preset threshold are located as the initial heart region; and

[0089] The convex envelope and the initial heart region are extracted to obtain a third image.

[0090] In some embodiments, performing rib removal on the third image to obtain a heart image may include:

[0091] The first pair of rib connection points are obtained according to the tangent points between the convex envelope and the initial heart region;

[0092] Correspondingly, in the third image, a second pair of rib connection points which are closest to the first pair of rib connection points and whose voxel values ​​are within a third threshold range are found;

[0093] determining the shortest path between the connection points of the second pair of ribs; and

[0094] The convex envelope and the shortest path and the area therebetween are removed in the third image to obtain a cardiac image.

[0095] As used herein, the term "computer" may include any processor-based or microprocessor-based system, including systems using microcontrollers, reduced instruction set computers (RISC), application specific integrated circuits (ASICs), logic circuits, and any other circuits or processors capable of performing the functions described herein. The above examples are exemplary only, and are not intended to limit the definition and / or meaning of the term "computer" in any way.

[0096] The instruction set may include various commands that instruct a computer or processor as a processing machine to perform specific operations, such as the methods and processes of various embodiments. The instruction set may take the form of a software program that may form part of one or more tangible non-transitory computer-readable media. The software may take various forms such as system software or application software. In addition, the software may take the form of a collection of independent programs or modules, a program module within a larger program, or a portion of a program module. The software may also include modular programming in the form of object-oriented programming. The processing of input data by the processing machine may be in response to operator commands, or in response to previous processing results, or in response to a request made by another processing machine.

[0097] Some exemplary embodiments have been described above, however, it should be understood that various modifications may be made. For example, if the described techniques are performed in a different order and / or if the components in the described systems, architectures, devices, or circuits are combined in different ways and / or replaced or supplemented by other components or their equivalents, suitable results may be achieved. Accordingly, other implementations also fall within the scope of protection of the claims.

Claims

1. A method for processing a cardiac CT image, comprising: reconstructing a cardiac CT image of the subject based on projection data of a cardiac region of the subject; Processing voxel points in the cardiac CT image whose voxel values ​​are within a first threshold range to obtain a first image, wherein the first threshold range is defined by the CT value of the metal object; Performing lung region segmentation on the first image to obtain a second image; performing initial heart region segmentation on the second image to obtain a third image; as well as The third image is subjected to rib removal to obtain a heart image.

2. The method of claim 1, wherein: Processing voxel points in the cardiac CT image whose voxel values ​​are within a first threshold range includes replacing the voxel values ​​within the first threshold range with preset voxel values.

3. The method of claim 1, wherein: Further including: It is determined whether there is a metal object in the heart of the detected object based on the voxel value in the heart CT image.

4. The method of claim 1, wherein: The lung region segmentation includes: Identify a region of voxel points in the first image whose voxel values ​​are within a second threshold range as a lung region, and identify a convex envelope of the lung region; and The convex envelope and its internal area are extracted to obtain the second image.

5. The method of claim 4, wherein: The initial heart region segmentation includes: Based on the second image, calculating the distance between each voxel point not in the lung region and all voxel points in the lung region to obtain the shortest distance corresponding to each voxel point; Identify the region where the corresponding voxel points whose shortest distance is greater than or equal to a preset threshold are located as the initial heart region; and The convex envelope and the initial heart region are extracted to obtain the third image.

6. The method of claim 5, wherein: The preset threshold is a product of a maximum value of the shortest distances corresponding to voxel points not in the lung region and a predetermined proportionality coefficient.

7. The method of claim 5, wherein: The rib removal includes: Obtaining a first pair of rib connection points according to the tangent points between the convex envelope and the initial heart region; Correspondingly finding in the third image a second pair of rib connection points which are closest to the first pair of rib connection points and whose voxel values ​​are within a third threshold range; determining the shortest path between the second pair of rib connection points; and The convex envelope and the shortest path and the area therebetween are removed from the third image to obtain the heart image.

8. A non-transitory computer-readable storage medium for storing a computer program, wherein when the computer program is executed by a computer, the computer executes the method for processing cardiac CT images according to any one of claims 1 to 7.

9. A cardiac CT image processing device, comprising: An image reconstruction module, configured to reconstruct a cardiac CT image of the subject based on projection data of a cardiac region of the subject; A preprocessing module, configured to process voxel points in the cardiac CT image whose voxel values ​​are within a first threshold range and obtain a first image, wherein the first threshold range is defined by the CT value of the metal object; A lung segmentation module, configured to perform lung region segmentation on the first image to obtain a second image; an initial segmentation module, configured to perform initial cardiac region segmentation on the second image to obtain a third image; as well as A rib removal module is configured to perform rib removal on the third image to obtain a heart image.

10. The device of claim 9, wherein: The lung segmentation module comprises: a lung recognition module configured to recognize a region of voxel points in the first image whose voxel values ​​are within a second threshold range as a lung region, and to recognize a convex envelope of the lung region; and A first extraction module is configured to extract the convex envelope and its internal area to obtain the second image.

11. The device of claim 10, wherein: The initial segmentation module comprises: a distance calculation module, configured to calculate the distance between each voxel point not in the lung region and all voxel points in the lung region based on the second image and obtain the shortest distance corresponding to each voxel point; A heart recognition module configured to recognize a region where voxel points whose corresponding shortest distance is greater than or equal to a preset threshold are located as an initial heart region; and A second extraction module is configured to extract the convex envelope and the initial heart region to obtain the third image.

12. The device of claim 11, wherein: The rib removal module comprises: a connection point determination module, configured to obtain a first pair of rib connection points according to the tangent points between the convex envelope and the initial heart region, and to correspondingly find in the third image a second pair of rib connection points which are closest to the first pair of rib connection points and whose voxel values ​​are within a third threshold range; a rib identification module configured to determine a shortest path between the second pair of rib connection points; and A third extraction module is configured to remove the convex envelope and the shortest path and the area therebetween in the third image to obtain the heart image.

Images