Cardiac image three-dimensional reconstruction method, device, equipment and storage medium

Abstract

The present invention belongs to the field of computer technology, and discloses a method, device, equipment and storage medium for three-dimensional reconstruction of a heart image. The present invention obtains three-dimensional contour points by performing image segmentation on an input heart image; selects a number of feature points in the heart image, and determines the initial shape of the heart corresponding to the heart image according to the feature points; and performs three-dimensional reconstruction based on the initial shape of the heart and the three-dimensional contour points. Since the heart image is first segmented to obtain three-dimensional contour points, and then feature points are selected in the heart image, the initial shape of the heart is obtained by transforming the selected feature points, and finally the initial shape of the heart is combined with the three-dimensional contour points for three-dimensional modeling, the accuracy of three-dimensional modeling can be improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method, device, equipment and storage medium for three-dimensional reconstruction of cardiac images. Background Art

[0002] Clinically, the evaluation of cardiac ejection fraction and myocardial mass, as well as other functional parameters (such as ventricular wall motion and wall thickness), is one of the important means of early diagnosis of heart disease. However, the heart is a complex three-dimensional organ with contraction and relaxation movements, and clinically, cardiac motion parameters play a more obvious role in describing its local abnormalities and early microlesions. This requires accurate segmentation of the three-dimensional structure of cardiac images in different phases (at different times in diastole and systole), and three-dimensional modeling based on the segmentation results to obtain accurate static and dynamic parameters of the heart. However, with the substantial improvement in the temporal and spatial resolution of imaging equipment, the massive amount of image data has greatly increased the difficulty of segmentation.

[0003] The above contents are only used to assist in understanding the technical solution of the present invention and do not constitute an admission that the above contents are prior art. Summary of the invention

[0004] The main purpose of the present invention is to provide a method, device, equipment and storage medium for three-dimensional reconstruction of cardiac images, aiming to solve the technical problems that the existing technology is difficult to accurately segment cardiac images and difficult to three-dimensional model.

[0005] To achieve the above object, the present invention provides a method for three-dimensional reconstruction of cardiac images, the method comprising the following steps:

[0006] Perform image segmentation on the input heart image to obtain three-dimensional contour points;

[0007] Selecting a plurality of feature points in the heart image, and determining an initial heart shape corresponding to the heart image according to the feature points;

[0008] Three-dimensional reconstruction is performed based on the initial heart shape and the three-dimensional contour points.

[0009] Optionally, the step of selecting a plurality of feature points in the heart image and determining the initial heart shape corresponding to the heart image according to the feature points includes:

[0010] Selecting outflow tract feature points, middle layer feature points and apex feature points in the heart image;

[0011] The initial shape of the heart corresponding to the heart image is determined through a preset point distribution model based on the outflow tract feature points, the middle layer feature points and the apex feature points.

[0012] Optionally, the step of selecting outflow tract feature points, middle layer feature points and apex feature points in the cardiac image includes:

[0013] Acquire a first outflow tract marker image, a second outflow tract marker image, an intermediate layer marker image, and an apex marker image;

[0014] selecting outflow tract feature points in the heart image according to the first outflow tract marking image and the second outflow tract marking image;

[0015] Selecting intermediate layer feature points in the heart image according to the intermediate layer marker image;

[0016] A cardiac apex feature point is selected in the cardiac image according to the cardiac apex marker image.

[0017] Optionally, the step of selecting outflow tract feature points in the heart image according to the first outflow tract marking image and the second outflow tract marking image includes:

[0018] Performing image registration on the first outflow tract marker image and the second outflow tract marker image with the heart image to obtain a first registered image and a second registered image;

[0019] Performing similarity matching between the first registered image and the heart image to obtain a first similarity;

[0020] Performing similarity matching between the second registered image and the heart image to obtain a second similarity;

[0021] The first similarity and the second similarity are compared, and outflow tract feature points are selected in the heart image according to the comparison result.

[0022] Optionally, the step of comparing the first similarity and the second similarity and selecting outflow tract feature points in the heart image according to the comparison result includes:

[0023] comparing the first similarity with the second similarity;

[0024] If the first similarity is greater than the second similarity, selecting outflow tract feature points in the heart image according to the first outflow tract marker image;

[0025] If the second similarity is greater than the first similarity, outflow tract feature points are selected in the heart image according to the second outflow tract marker image.

[0026] Optionally, the step of performing three-dimensional reconstruction based on the initial heart shape and the three-dimensional contour points includes:

[0027] establishing a coordinate system based on the cardiac image;

[0028] Determine the cardiac bottom layer contour points according to the three-dimensional contour points, and adjust the cardiac image so that the plane formed by the cardiac bottom layer contour points is located in the horizontal direction;

[0029] Adjusting the initial shape of the heart so that the heart layer in the initial shape of the heart is located in a horizontal direction;

[0030] The initial shape of the heart is mapped to a coordinate system in the heart image to obtain a three-dimensional reconstructed image.

[0031] Optionally, the step of mapping the initial heart shape to a coordinate system in the heart image to obtain a three-dimensional reconstructed image includes:

[0032] Determine the cardiac base layer and the cardiac apex layer in the cardiac image based on the three-dimensional contour points;

[0033] Vertically stretching the left and right ventricles in the initial heart shape so that the basal layer and apical layer in the initial heart shape cover the basal layer and apical layer in the heart image, thereby obtaining a stretched initial heart shape;

[0034] The stretched initial shape of the heart is mapped to a coordinate system in the heart image to obtain a three-dimensional reconstructed image.

[0035] In addition, to achieve the above-mentioned purpose, the present invention also proposes a three-dimensional reconstruction device for cardiac images, which includes the following modules:

[0036] An image segmentation module is used to perform image segmentation on an input heart image to obtain three-dimensional contour points;

[0037] A shape determination module, used for selecting a plurality of feature points in the heart image, and determining an initial heart shape corresponding to the heart image according to the feature points;

[0038] A three-dimensional reconstruction module is used to perform three-dimensional reconstruction based on the initial shape of the heart and the three-dimensional contour points.

[0039] In addition, to achieve the above-mentioned purpose, the present invention also proposes a three-dimensional reconstruction device for cardiac images, which includes: a processor, a memory, and a three-dimensional reconstruction program for cardiac images stored in the memory and executable on the processor, and the three-dimensional reconstruction program for cardiac images implements the steps of the three-dimensional reconstruction method for cardiac images as described above when executed by the processor.

[0040] In addition, to achieve the above objectives, the present invention also proposes a computer-readable storage medium, on which a cardiac image three-dimensional reconstruction program is stored, and when the cardiac image three-dimensional reconstruction program is executed, the steps of the cardiac image three-dimensional reconstruction method as described above are implemented.

[0041] The present invention obtains three-dimensional contour points by performing image segmentation on the input heart image; selects a number of feature points in the heart image, and determines the initial heart shape corresponding to the heart image according to the feature points; and performs three-dimensional reconstruction based on the initial heart shape and the three-dimensional contour points. Since the heart image is first segmented to obtain three-dimensional contour points, then feature points are selected in the heart image, the initial heart shape is obtained by transforming the selected feature points, and finally the initial heart shape is combined with the three-dimensional contour points for three-dimensional modeling, the accuracy of three-dimensional modeling can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a schematic diagram of the structure of an electronic device in a hardware operating environment involved in an embodiment of the present invention;

[0043] Figure 2 A schematic diagram of the flow chart of the first embodiment of the method for three-dimensional reconstruction of cardiac images of the present invention;

[0044] Figure 3 A schematic diagram of the flow chart of a second embodiment of a method for three-dimensional reconstruction of cardiac images according to the present invention;

[0045] Figure 4 FIG. 4 is a structural block diagram of a first embodiment of a device for 3D reconstruction of cardiac images according to the present invention.

[0046] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0047] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0048] Reference Figure 1 , Figure 1 The figure is a schematic diagram of the structure of a cardiac image three-dimensional reconstruction device in the hardware operating environment involved in the embodiment of the present invention.

[0049] like Figure 1As shown, the electronic device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (Wireless-Fidelity, WI-FI) interface). The memory 1005 may be a high-speed random access memory (Random Access Memory, RAM), or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk storage. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0050] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation on the electronic device, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0051] like Figure 1 As shown, the memory 1005 as a storage medium may include an operating system, a network communication module, a user interface module, and a cardiac image three-dimensional reconstruction program.

[0052] exist Figure 1 In the electronic device shown, the network interface 1004 is mainly used for data communication with a network server; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 in the electronic device of the present invention can be set in a cardiac image three-dimensional reconstruction device, and the electronic device calls the cardiac image three-dimensional reconstruction program stored in the memory 1005 through the processor 1001, and executes the cardiac image three-dimensional reconstruction method provided in an embodiment of the present invention.

[0053] The embodiment of the present invention provides a method for three-dimensional reconstruction of a cardiac image, referring to Figure 2 , Figure 2 FIG. 4 is a flow chart of a first embodiment of a method for three-dimensional reconstruction of cardiac images according to the present invention.

[0054] In this embodiment, the method for three-dimensional reconstruction of cardiac images includes the following steps:

[0055] Step S10: performing image segmentation on the input heart image to obtain three-dimensional contour points.

[0056] It should be noted that the executor of this embodiment can be the three-dimensional reconstruction device of the heart image, and the three-dimensional reconstruction device of the heart image can be an electronic device such as a personal computer or a server, or a device that can achieve the same or similar functions. This embodiment does not limit this. In this embodiment and the following embodiments, the three-dimensional reconstruction of the heart image is taken as an example to illustrate the three-dimensional reconstruction method of the heart image of the present invention.

[0057] It should be noted that the heart image can be a short-axis heart image, and the heart image can be manually input by the user, or can be collected and input by an image acquisition device, which is not limited in this embodiment, wherein the image acquisition device can be a nuclear magnetic resonance imaging device. The input heart image can be multiple, and can include heart images at different phases. Performing image segmentation on the input heart image to obtain three-dimensional contour points can be performing image segmentation on the input heart image through a preset image segmentation model, determining the division of each heart region in the heart image, and extracting contour points of each heart region to obtain three-dimensional contour points, for example: segmenting the input heart image through a preset image segmentation model, dividing the left and right ventricle regions of the heart, and then extracting three-dimensional contour points based on the left and right ventricle regions of the heart. Among them, the preset image segmentation model can be a neural network model for image segmentation that has been pre-trained using a large number of samples.

[0058] Step S20: Selecting a plurality of feature points in the heart image, and determining an initial heart shape corresponding to the heart image according to the feature points.

[0059] It should be noted that the feature points may be feature points of different areas of the heart. After acquiring the feature points of the heart image at different phases, the average shape of the heart may be acquired. The average shape may then be subjected to composite transformations such as stretching and rotation based on the acquired feature points of the heart to obtain the initial shape of the heart. The average shape of the heart may be the shape information of the heart set in advance, and different shapes may be derived based on the average shape.

[0060] Further, in order to improve the efficiency of determining the initial shape of the heart, step S20 in this embodiment may include:

[0061] Selecting outflow tract feature points, middle layer feature points and apex feature points in the heart image;

[0062] The initial shape of the heart corresponding to the heart image is determined through a preset point distribution model based on the outflow tract feature points, the middle layer feature points and the apex feature points.

[0063] It should be noted that the preset point distribution model may be a pre-trained neural network model or a deep learning model. The outflow tract feature points may be feature points selected from the outflow tract region of the left ventricle of the heart, the middle layer feature points may be feature points selected from the middle layer of the right ventricle of the heart, and the apex feature points may be feature points selected from the apex regions of the left and right ventricles of the heart.

[0064] In actual use, the preset point distribution model pre-sets a number of feature points. It only needs to determine the position of each feature point in the heart image. The preset point distribution model can then perform a compound transformation on the average shape of the heart according to the marked feature point positions to obtain the initial shape of the heart. The average shape of the heart can be generated when the point distribution model is trained using a large number of samples.

[0065] For example, the feature points set by the preset point distribution model are five feature points, namely, the aorta (AORTA), the mitral valve (MITRAL), the left ventricular apex (LVAPEX), the tricuspid valve (TRICUPSID), and the right ventricular apex (RVAPEX). At this time, the outflow tract feature points may include the aorta (AORTA) and the mitral valve (MITRAL) feature points, the middle layer feature points may include the tricuspid valve (TRICUPSID) feature points, and the apex feature points may include the left ventricular apex (LVAPEX) and the right ventricular apex (RVAPEX) feature points.

[0066] Furthermore, due to differences in imaging environments, the left ventricular outflow tract image may be missing in the cardiac short-axis sequence images. In order to ensure that each feature point can be accurately selected, the step of selecting the outflow tract feature point, the middle layer feature point and the apex feature point in the cardiac image in this embodiment may include:

[0067] Acquire a first outflow tract marker image, a second outflow tract marker image, an intermediate layer marker image, and an apex marker image;

[0068] selecting outflow tract feature points in the heart image according to the first outflow tract marking image and the second outflow tract marking image;

[0069] Selecting intermediate layer feature points in the heart image according to the intermediate layer marker image;

[0070] A cardiac apex feature point is selected in the cardiac image according to the cardiac apex marker image.

[0071] It should be noted that the first outflow tract marked image may be a short-axis cardiac subcardiac image containing the left ventricular outflow tract, in which there are outflow tract feature points marked by experts. The second outflow tract marked image may be a short-axis cardiac subcardiac image without the left ventricular outflow tract, in which there are outflow tract feature points marked by experts. The middle layer marked image may be a short-axis cardiac middle layer image, in which there are middle layer feature points marked by experts. The apex marked image may be a short-axis cardiac subcardiac image, in which there are apex layer feature points marked by experts.

[0072] In actual use, selecting outflow tract feature points in the cardiac image based on the first outflow tract marker image and the second outflow tract marker image may be to compare the cardiac image with the first outflow tract marker image and the second outflow tract marker image respectively for similarity, select the first outflow tract marker image and the second outflow tract marker image with higher similarity as the target marker image according to the comparison result, and select outflow tract feature points in the cardiac image according to the target marker image. Selecting feature points in the cardiac image based on the marker image may be to convert the feature points marked by the expert in the marker image into feature points in the cardiac image by using a label transfer method, wherein the marker image may be any one of the first outflow tract marker image, the second outflow tract marker image, the intermediate layer marker image, and the apex marker image.

[0073] Furthermore, in order to ensure that the selected outflow tract feature points are as accurate as possible, the step of selecting the outflow tract feature points in the heart image according to the first outflow tract marker image and the second outflow tract marker image in this embodiment may include:

[0074] Performing image registration on the first outflow tract marker image and the second outflow tract marker image with the heart image to obtain a first registered image and a second registered image;

[0075] Performing similarity matching between the first registered image and the heart image to obtain a first similarity;

[0076] Performing similarity matching between the second registered image and the heart image to obtain a second similarity;

[0077] The first similarity and the second similarity are compared, and outflow tract feature points are selected in the heart image according to the comparison result.

[0078] It should be noted that, performing image registration on the first outflow tract marker image and the second outflow tract marker image with the heart image respectively to obtain the first registered image and the second registered image can be performed by performing image registration on the first outflow tract marker image and the heart image through a preset image registration algorithm to obtain the first registered image, and performing image registration on the second outflow tract marker image and the heart image through a preset image registration algorithm to obtain the second registered image, wherein the preset image registration algorithm can be a rigid registration algorithm or a non-rigid registration algorithm, which is not limited in the present embodiment.

[0079] The first registration image is matched with the heart image for similarity, and the first similarity can be obtained by matching the first registration image with the heart image for similarity using a preset similarity index to obtain the first similarity. The second registration image is matched with the heart image for similarity, and the second similarity can be matched with the second registration image with the heart image for similarity using a preset similarity index to obtain the second similarity. The preset similarity metric can be an indicator such as maximum mutual information or AM (Alignment Metric) measurement.

[0080] In actual use, comparing the first similarity and the second similarity, and selecting the outflow tract feature point in the heart image according to the comparison result can be performed by comparing the first similarity and the second similarity and generating a comparison result, selecting a target mark image from the first outflow tract mark image and the second outflow tract mark image according to the comparison result, and then selecting the outflow tract feature point in the heart image according to the target mark image.

[0081] It can be understood that if the comparison result is that the first similarity is greater than the second similarity, it means that the similarity between the first registered image and the heart image is higher than the similarity between the second registered image and the heart image, which means that the heart image contains the left ventricular outflow tract image, and therefore, the first outflow tract marker image can be used as the target marker image, and outflow tract feature points can be selected in the heart image according to the first outflow tract marker image. Conversely, if the comparison result is that the second similarity is greater than the first similarity, it means that the heart image does not contain the left ventricular outflow tract image, and therefore, the second outflow tract marker image can be used as the target marker image, and outflow tract feature points can be selected in the heart image according to the second outflow tract marker image.

[0082] Step S30: performing three-dimensional reconstruction based on the initial heart shape and the three-dimensional contour points.

[0083] It is understandable that after the image is segmented to determine the three-dimensional contour points of each heart region in the heart image, it is only necessary to map the obtained initial shape of the heart to the three-dimensional contour points to complete the three-dimensional reconstruction and obtain the three-dimensional modeling result of the heart.

[0084] It is understandable that the heart is a moving organ, and the 3D modeling results of the heart in all phases can be obtained by simply assigning the initial shape of the second phase to the 3D modeling results of the heart by segmenting the heart image in the first phase, and so on.

[0085] This embodiment obtains three-dimensional contour points by performing image segmentation on the input heart image; selects a number of feature points in the heart image, and determines the initial heart shape corresponding to the heart image based on the feature points; and performs three-dimensional reconstruction based on the initial heart shape and the three-dimensional contour points. Since the heart image is first segmented to obtain three-dimensional contour points, then feature points are selected in the heart image, the initial heart shape is obtained by transforming the selected feature points, and finally the initial heart shape is combined with the three-dimensional contour points for three-dimensional modeling, the accuracy of three-dimensional modeling can be improved.

[0086] refer to Figure 3 , Figure 3 FIG. 4 is a flow chart of a second embodiment of a method for three-dimensional reconstruction of cardiac images according to the present invention.

[0087] Based on the first embodiment, step S30 of the method for 3D reconstruction of cardiac images in this embodiment includes:

[0088] Step S301: establishing a coordinate system based on the cardiac image.

[0089] It should be noted that establishing a coordinate system based on a cardiac image may be to select any point in the cardiac image as the origin to establish the coordinate system.

[0090] Step S302: determining the cardiac bottom layer contour points according to the three-dimensional contour points, and adjusting the cardiac image so that the plane formed by the cardiac bottom layer contour points is located in the horizontal direction.

[0091] It should be noted that, determining the cardiac bottom layer contour points based on the three-dimensional contour points and adjusting the cardiac image so that the plane formed by the cardiac bottom layer contour points is located in the horizontal direction can be by screening the cardiac bottom layer contour points from the three-dimensional contour points, determining the plane formed by the cardiac bottom layer contour points, and rotating the cardiac image so that the cardiac bottom layer contour points are shrunk into a plane that is located in the horizontal opposite direction in the constructed coordinate system.

[0092] Step S303: adjusting the initial shape of the heart so that the cardiac stratum in the initial shape of the heart is located in a horizontal direction.

[0093] It is understandable that adjusting the initial shape of the heart may also be rotating the initial shape of the heart so that the bottom layer of the heart in the initial shape of the heart is located in a horizontal direction.

[0094] Step S304: Mapping the initial shape of the heart to the coordinate system in the heart image to obtain a three-dimensional reconstructed image.

[0095] It should be noted that mapping the initial shape of the heart to the coordinate system of the heart image to obtain a three-dimensional reconstructed image can be achieved by matching the initial shape of the heart with the heart area in the heart image one by one, inversely transforming the initial shape of the heart, transforming it to the coordinate system in the heart image, and then segmenting the initial shape of the heart according to the three-dimensional contour points to obtain a three-dimensional reconstructed image.

[0096] Furthermore, in the actual process, the constructed initial heart shape is often difficult to effectively cover the cardiac basal layer and the cardiac apex layer in the cardiac image. In order to ensure the effect of three-dimensional reconstruction, step S304 in this embodiment may include:

[0097] Determine the cardiac base layer and the cardiac apex layer in the cardiac image based on the three-dimensional contour points;

[0098] Vertically stretching the left and right ventricles in the initial heart shape so that the basal layer and apical layer in the initial heart shape cover the basal layer and apical layer in the heart image, thereby obtaining a stretched initial heart shape;

[0099] The stretched initial shape of the heart is mapped to a coordinate system in the heart image to obtain a three-dimensional reconstructed image.

[0100] It can be understood that by vertically stretching the left and right ventricles in the initial shape of the heart and making the base layer and apex layer in the initial shape of the heart cover the base layer and apex layer in the heart image, it can be ensured that the stretched initial shape of the heart can effectively cover the heart in the heart image, thereby ensuring that the initial shape of the heart can be accurately segmented based on the three-dimensional contour points to obtain an accurate three-dimensional reconstructed image.

[0101] This embodiment establishes a coordinate system based on the heart image; determines the contour points of the heart layer according to the three-dimensional contour points, adjusts the heart image so that the plane formed by the contour points of the heart layer is in the horizontal direction; adjusts the initial shape of the heart so that the heart layer in the initial shape of the heart is in the horizontal direction; maps the initial shape of the heart to the coordinate system in the heart image to obtain a three-dimensional reconstructed image. Because before three-dimensional modeling is performed based on the three-dimensional contour points and the initial shape of the heart, the heart image and the initial shape of the heart are adjusted with the heart layer as a reference, so that the heart layer in the heart image and the initial shape of the heart are located in the same horizontal plane before three-dimensional modeling is performed, the accuracy of three-dimensional modeling is further improved.

[0102] In addition, an embodiment of the present invention further provides a storage medium on which a three-dimensional reconstruction program for cardiac images is stored. When the three-dimensional reconstruction program for cardiac images is executed by a processor, the steps of the three-dimensional reconstruction method for cardiac images described above are implemented.

[0103] Reference Figure 4 , Figure 4 FIG. 4 is a structural block diagram of a first embodiment of a device for 3D reconstruction of cardiac images according to the present invention.

[0104] like Figure 4 As shown, the cardiac image three-dimensional reconstruction device proposed in the embodiment of the present invention includes:

[0105] An image segmentation module 10 is used to perform image segmentation on an input heart image to obtain three-dimensional contour points;

[0106] A shape determination module 20, configured to select a plurality of feature points in the heart image and determine an initial heart shape corresponding to the heart image according to the feature points;

[0107] The three-dimensional reconstruction module 30 is used to perform three-dimensional reconstruction based on the initial shape of the heart and the three-dimensional contour points.

[0108] This embodiment obtains three-dimensional contour points by performing image segmentation on the input heart image; selects a number of feature points in the heart image, and determines the initial heart shape corresponding to the heart image based on the feature points; and performs three-dimensional reconstruction based on the initial heart shape and the three-dimensional contour points. Since the heart image is first segmented to obtain three-dimensional contour points, then feature points are selected in the heart image, the initial heart shape is obtained by transforming the selected feature points, and finally the initial heart shape is combined with the three-dimensional contour points for three-dimensional modeling, the accuracy of three-dimensional modeling can be improved.

[0109] Furthermore, the shape determination module 20 is also used to select outflow tract feature points, middle layer feature points and apex feature points in the heart image; and determine the initial heart shape corresponding to the heart image through a preset point distribution model based on the outflow tract feature points, the middle layer feature points and the apex feature points.

[0110] Furthermore, the shape determination module 20 is also used to obtain a first outflow tract marking image, a second outflow tract marking image, an intermediate layer marking image and an apex marking image; select outflow tract feature points in the heart image according to the first outflow tract marking image and the second outflow tract marking image; select intermediate layer feature points in the heart image according to the intermediate layer marking image; select apex feature points in the heart image according to the apex marking image.

[0111] Furthermore, the shape determination module 20 is also used to perform image registration on the first outflow tract marker image and the second outflow tract marker image with the heart image respectively to obtain a first registered image and a second registered image; perform similarity matching on the first registered image and the heart image to obtain a first similarity; perform similarity matching on the second registered image and the heart image to obtain a second similarity; compare the first similarity and the second similarity, and select outflow tract feature points in the heart image according to the comparison result.

[0112] Furthermore, the shape determination module 20 is also used to compare the first similarity with the second similarity; if the first similarity is greater than the second similarity, outflow tract feature points are selected in the heart image according to the first outflow tract marker image; if the second similarity is greater than the first similarity, outflow tract feature points are selected in the heart image according to the second outflow tract marker image.

[0113] Furthermore, the three-dimensional reconstruction module 30 is also used to establish a coordinate system based on the heart image; determine the heart bottom layer contour points according to the three-dimensional contour points, and adjust the heart image so that the plane formed by the heart bottom layer contour points is located in the horizontal direction; adjust the initial shape of the heart so that the heart bottom layer in the initial shape of the heart is located in the horizontal direction; map the initial shape of the heart to the coordinate system in the heart image to obtain a three-dimensional reconstructed image.

[0114] Furthermore, the three-dimensional reconstruction module 30 is also used to determine the base layer and apex layer in the heart image based on the three-dimensional contour points; vertically stretch the left and right ventricles in the initial shape of the heart so that the base layer and apex layer in the initial shape of the heart cover the base layer and apex layer in the heart image to obtain the stretched initial shape of the heart; map the stretched initial shape of the heart to the coordinate system in the heart image to obtain a three-dimensional reconstructed image.

[0115] It should be understood that the above is only an example and does not constitute any limitation on the technical solution of the present invention. In specific applications, technicians in this field can make settings as needed, and the present invention does not limit this.

[0116] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of the present invention. In practical applications, technicians in this field can select part or all of them according to actual needs to achieve the purpose of the present embodiment, and no limitation is made here.

[0117] In addition, for technical details not fully described in this embodiment, reference may be made to the three-dimensional reconstruction method of cardiac images provided in any embodiment of the present invention, and will not be repeated here.

[0118] In addition, it should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0119] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0120] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory (ROM) / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0121] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for three-dimensional reconstruction of cardiac images, characterized in that: The method for three-dimensional reconstruction of cardiac images comprises the following steps: Perform image segmentation on the input heart image to obtain three-dimensional contour points; Selecting a plurality of feature points in the heart image, and determining an initial heart shape corresponding to the heart image according to the feature points; Performing three-dimensional reconstruction based on the initial shape of the heart and the three-dimensional contour points; The step of selecting a plurality of feature points from the heart image and determining the initial heart shape corresponding to the heart image according to the feature points comprises: Selecting outflow tract feature points, middle layer feature points and apex feature points in the heart image; Determine the initial heart shape corresponding to the heart image through a preset point distribution model based on the outflow tract feature points, the middle layer feature points and the apex feature points; The step of selecting outflow tract feature points, middle layer feature points and apex feature points in the heart image comprises: Acquire a first outflow tract marker image, a second outflow tract marker image, an intermediate layer marker image, and an apex marker image; selecting outflow tract feature points in the heart image according to the first outflow tract marking image and the second outflow tract marking image; Selecting intermediate layer feature points in the heart image according to the intermediate layer marker image; Selecting a cardiac apex feature point in the cardiac image according to the cardiac apex marker image; The step of selecting outflow tract feature points in the heart image according to the first outflow tract marking image and the second outflow tract marking image comprises: Performing image registration on the first outflow tract marker image and the second outflow tract marker image with the heart image to obtain a first registered image and a second registered image; Performing similarity matching between the first registered image and the heart image to obtain a first similarity; Performing similarity matching between the second registered image and the heart image to obtain a second similarity; comparing the first similarity and the second similarity, and selecting an outflow tract feature point in the heart image according to the comparison result; The step of comparing the first similarity and the second similarity and selecting outflow tract feature points in the heart image according to the comparison result includes: comparing the first similarity with the second similarity; If the first similarity is greater than the second similarity, selecting outflow tract feature points in the heart image according to the first outflow tract marker image; If the second similarity is greater than the first similarity, outflow tract feature points are selected in the heart image according to the second outflow tract marker image.

2. The method for three-dimensional reconstruction of cardiac images according to claim 1, characterized in that: The step of performing three-dimensional reconstruction based on the initial heart shape and the three-dimensional contour points comprises: establishing a coordinate system based on the cardiac image; Determine the cardiac bottom layer contour points according to the three-dimensional contour points, and adjust the cardiac image so that the plane formed by the cardiac bottom layer contour points is located in the horizontal direction; Adjusting the initial shape of the heart so that the heart layer in the initial shape of the heart is located in a horizontal direction; The initial shape of the heart is mapped to a coordinate system in the heart image to obtain a three-dimensional reconstructed image.

3. The method for three-dimensional reconstruction of cardiac images according to claim 2, characterized in that: The step of mapping the initial heart shape to the coordinate system in the heart image to obtain a three-dimensional reconstructed image comprises: Determine the cardiac base layer and the cardiac apex layer in the cardiac image based on the three-dimensional contour points; Vertically stretching the left and right ventricles in the initial heart shape so that the basal layer and apical layer in the initial heart shape cover the basal layer and apical layer in the heart image, thereby obtaining the stretched initial heart shape; The stretched initial shape of the heart is mapped to a coordinate system in the heart image to obtain a three-dimensional reconstructed image.

4. A three-dimensional reconstruction device for cardiac images, characterized in that: The cardiac image three-dimensional reconstruction device comprises the following modules: An image segmentation module is used to perform image segmentation on an input heart image to obtain three-dimensional contour points; A shape determination module, used for selecting a plurality of feature points in the heart image, and determining an initial heart shape corresponding to the heart image according to the feature points; A three-dimensional reconstruction module, used for performing three-dimensional reconstruction based on the initial shape of the heart and the three-dimensional contour points; The three-dimensional reconstruction module is further used to select outflow tract feature points, middle layer feature points and apex feature points in the heart image; and determine the initial heart shape corresponding to the heart image through a preset point distribution model based on the outflow tract feature points, the middle layer feature points and the apex feature points; The three-dimensional reconstruction module is further used to obtain a first outflow tract marker image, a second outflow tract marker image, an intermediate layer marker image and an apex marker image; select outflow tract feature points in the heart image according to the first outflow tract marker image and the second outflow tract marker image; select intermediate layer feature points in the heart image according to the intermediate layer marker image; select apex feature points in the heart image according to the apex marker image; The three-dimensional reconstruction module is further used to perform image registration on the first outflow tract marker image and the second outflow tract marker image with the heart image respectively to obtain a first registered image and a second registered image; perform similarity matching on the first registered image and the heart image to obtain a first similarity; perform similarity matching on the second registered image and the heart image to obtain a second similarity; compare the first similarity and the second similarity, and select outflow tract feature points in the heart image according to the comparison result; The three-dimensional reconstruction module is further used to compare the first similarity with the second similarity; if the first similarity is greater than the second similarity, outflow tract feature points are selected in the heart image according to the first outflow tract marker image; if the second similarity is greater than the first similarity, outflow tract feature points are selected in the heart image according to the second outflow tract marker image.

5. A cardiac image three-dimensional reconstruction device, characterized in that: The three-dimensional reconstruction device of cardiac images includes: a processor, a memory, and a three-dimensional reconstruction program of cardiac images stored in the memory and executable on the processor. When the three-dimensional reconstruction program of cardiac images is executed by the processor, the steps of the three-dimensional reconstruction method of cardiac images as described in any one of claims 1 to 3 are implemented.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a three-dimensional cardiac image reconstruction program, which, when executed, implements the steps of the three-dimensional cardiac image reconstruction method according to any one of claims 1 to 3.

Images