Three-dimensional body construction method and device based on CT image, storage medium and terminal

Abstract

The application discloses a three-dimensional body construction method and device based on CT images, a storage medium and a terminal. A CT sequence image of a target object is acquired. In response to a user's manual contour segmentation operation on a first target CT image in the CT sequence image, the first target CT image is subjected to first contour segmentation, and a first contour image of the target object is obtained. Based on each first contour image, a second target CT image in the CT sequence image, except for the first target CT image, is subjected to second contour segmentation, and a second contour image of the target object is obtained. A three-dimensional body of the target object is constructed according to each first contour image and each second contour image. Since the second contour segmentation is active processing and calculation on the manually segmented first contour image, when the second contour image is obtained, manual operation on a large number of images can be avoided, human-computer interaction is reduced, the three-dimensional body construction efficiency is improved, and the accuracy of the three-dimensional body of the target object can be ensured by the manually segmented first contour image.

Description

Technical Field

[0001] This application relates to the field of computer image processing technology, and in particular to a method, apparatus, storage medium, and terminal for constructing three-dimensional volumes based on CT images. Background Technology

[0002] In scientific research in fields such as medicine, biology, and geology, the target object can be observed based on its CT images. In order to observe the internal structure of the target object from various angles, a three-dimensional volume is usually constructed based on the CT images of the target object to obtain a three-dimensional projection image that can be observed from any perspective. This makes it convenient for researchers to observe and diagnose the structure of the internal tissues or organs of the target object.

[0003] Constructing accurate 3D volumes can help medical personnel and researchers accurately judge and analyze the situation of target objects, which is conducive to the development of modern science. However, current 3D volume construction methods require complex procedures for handling complex cases. Therefore, there is an urgent need to develop a simple and accurate 3D volume construction method. Summary of the Invention

[0004] This application provides a method, apparatus, storage medium, and terminal for constructing three-dimensional volumes based on CT images, which can solve the technical problems of cumbersome interaction process and inaccurate three-dimensional construction results in related technologies.

[0005] In a first aspect, embodiments of this application provide a method for constructing a three-dimensional volume based on CT images, the method comprising:

[0006] Acquire CT sequence images of the target object; in response to a user's manual contour segmentation operation on a first target CT image in the CT sequence images, perform first contour segmentation on the first target CT image to obtain a first contour image of the target object.

[0007] Based on each first contour image, the second contour segmentation is actively performed on the second target CT image in the CT sequence image excluding the first target CT image to obtain the second contour image of the target object;

[0008] The three-dimensional volume of the target object is constructed based on each first contour image and each second contour image.

[0009] Secondly, embodiments of this application provide a three-dimensional volume construction device based on CT images, the device comprising:

[0010] The first segmentation module is used to acquire CT sequence images of the target object, and in response to a user's manual contour segmentation operation on the first target CT image in the CT sequence images, performs first contour segmentation on the first target CT image to obtain the first contour image of the target object.

[0011] The second segmentation module is used to actively perform second contour segmentation on the second target CT image (excluding the first target CT image) in the CT sequence image based on each first contour image, so as to obtain the second contour image of the target object.

[0012] A 3D volume construction module is used to construct a 3D volume of the target object based on each first contour image and each second contour image.

[0013] Thirdly, embodiments of this application provide a computer storage medium storing a plurality of instructions adapted for loading by a processor and executing the steps of the method described above.

[0014] Fourthly, embodiments of this application provide a terminal, including a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being adapted to be loaded by the processor and to execute the steps of the above-described method.

[0015] The beneficial effects of the technical solutions provided in some embodiments of this application include at least the following:

[0016] This application provides a method for constructing a 3D volume based on CT images. The method involves acquiring a CT sequence image of a target object, responding to a user's manual contour segmentation operation on a first target CT image within the CT sequence image, performing a first contour segmentation on the first target CT image to obtain a first contour image of the target object, and then actively performing a second contour segmentation on a second target CT image (excluding the first target CT image) within the CT sequence image based on each first contour image. Finally, a 3D volume of the target object is constructed based on each first contour image and each second contour image. Since the second contour segmentation is an active processing and calculation of the manually segmented first contour images, obtaining the second contour image avoids a large amount of manual operation on CT images, reduces human-computer interaction, accelerates the 3D volume construction efficiency of the target object, and ensures the accuracy of the 3D volume construction by using manually segmented first contour images. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 An exemplary system architecture diagram of a three-dimensional volume construction method based on CT images provided in this application embodiment;

[0019] Figure 2 A flowchart illustrating a three-dimensional volume construction method based on CT images provided in this application embodiment;

[0020] Figure 3 A flowchart illustrating a three-dimensional volume construction method based on CT images provided in this application embodiment;

[0021] Figure 4 A schematic diagram illustrating a process for segmenting target objects in CT sequence images, provided as an embodiment of this application;

[0022] Figure 5 A structural block diagram of a three-dimensional volume construction device based on CT images provided in this application embodiment;

[0023] Figure 6 This is a schematic diagram of the structure of a terminal provided in an embodiment of this application. Detailed Implementation

[0024] To make the features and advantages of this application more apparent and understandable, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0025] In the following description, when referring to the accompanying drawings, the same numbers in different drawings denote the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0026] In recent years, with the rapid development of computer technology and the increasing maturity of image and graphics technology, digital image processing and computer technology have been widely applied in fields such as medicine and biology, using digital image processing to observe the characteristics of objects. Relying on the development of modern medical imaging technology, medical personnel can obtain diagnostic images of human organs through various medical instruments. For example, X-ray computed tomography (X-CT) and ultrasound imaging can provide morphological information of human organs and tissues; nuclear medicine images such as SPECT, PET, and magnetic resonance imaging (MRI) not only provide morphological information of tissues and organs, clearly displaying structures such as cartilage, but also provide information on the distribution and bioactivity of relevant tracer elements, which helps in clinical diagnosis and scientific research. With the help of these technologies, doctors and researchers can make diagnoses more accurately and conveniently, greatly improving the accuracy of clinical diagnosis and having significant practical implications.

[0027] Traditional digital imaging techniques typically involve acquiring image data of a specific cross-section of the target object and then displaying it on film or a screen for analysis and diagnosis. However, whether displayed on film or a screen, observers can only view two-dimensional images and can only observe the images from a fixed orientation. In this case, observers rely primarily on experience to analyze the images, resulting in highly subjective diagnostic results. Furthermore, understanding complex or even deformed three-dimensional structures presents significant challenges. The application of 3D reconstruction technology can improve this situation. 3D reconstruction utilizes computer technology to segment and reconstruct the target object from acquired tomographic images, enabling observers to analyze the imaging data from multiple angles and levels, thereby assessing the state and condition of the target object. Therefore, the application of 3D reconstruction technology will significantly improve the efficiency and accuracy of target object observation.

[0028] The accuracy and efficiency of target object segmentation in CT sequence images directly determine the quality of 3D reconstruction technology, and there are currently various target object segmentation algorithms for 3D images. One segmentation method is region-based segmentation, which selects corresponding positions by repeatedly adjusting pixel thresholds. However, this method does not consider spatial information from different angles, cannot accurately construct complex 3D objects, and the interactive process and operation of repeatedly adjusting thresholds are quite cumbersome. Another segmentation method is edge-based segmentation, which attempts to detect possible boundaries in the image to complete the segmentation task by detecting pixels with drastic changes in pixel values. This method is relatively convenient to interact with, requiring only the initial outline to be drawn, but the quality of the initial outline affects the final result. There are also machine learning-based segmentation methods, mainly divided into unsupervised and supervised methods. Unsupervised methods mainly use clustering algorithms, including K-means and EM algorithms. These methods are simple to interact with, but their performance is generally average in complex segmentation tasks. Supervised methods, on the other hand, select a portion of already segmented pixels or images as a training set after extracting features from pixels to train a classification model, and then use this model to complete the image segmentation. This method requires a large number of accurately labeled training sets for training, which makes the preparation of samples and the training of the model more difficult.

[0029] In practice, preparing the training sample set for training automatic segmentation models is quite difficult, and automatic segmentation models generally perform poorly in handling complex cases. Furthermore, interactive segmentation based on user operations requires high user skill and involves cumbersome steps, which can lead to a slow segmentation process and waste a significant amount of user time and effort.

[0030] Therefore, this application provides a method for obtaining CT sequence images of a target object, manually performing first contour segmentation on a first target CT image in the CT sequence image to obtain a first contour image of the target object; actively performing second contour segmentation on the remaining second target CT images in the CT sequence image based on each first contour image to obtain a second contour image of the target object; and constructing a three-dimensional volume of the target object based on each first contour image and each second contour image, in order to solve the above-mentioned technical problems.

[0031] Please see Figure 1 , Figure 1 An exemplary system architecture diagram of a three-dimensional volume construction method based on CT images provided in this application embodiment.

[0032] like Figure 1As shown, the system architecture may include a terminal 101, a network 102, and a server 103. The network 102 serves as the medium for providing a communication link between the terminal 101 and the server 103. The network 102 may include various types of wired or wireless communication links, such as wired communication links including fiber optic cables, twisted-pair cables, or coaxial cables, and wireless communication links including Bluetooth communication links, Wireless-Fidelity (Wi-Fi) communication links, or microwave communication links, etc.

[0033] Terminal 101 can interact with server 103 via network 102 to receive messages from or send messages to server 103. Alternatively, terminal 101 can interact with server 103 via network 102 to receive messages or data sent to server 103 by other users. Terminal 101 can be hardware or software. When terminal 101 is hardware, it can be various electronic devices, including but not limited to smartwatches, smartphones, tablets, laptops, and desktop computers. When terminal 101 is software, it can be installed in the aforementioned electronic devices and can be implemented as multiple software programs or software modules (e.g., to provide distributed services) or as a single software program or software module; no specific limitation is made here.

[0034] In this embodiment, terminal 101 first acquires CT sequence images of the target object. In response to a user's manual contour segmentation operation on the first target CT image in the CT sequence images, terminal 101 performs first contour segmentation on the first target CT image to obtain a first contour image of the target object. Then, based on each first contour image, terminal 101 actively performs second contour segmentation on the second target CT image in the CT sequence images excluding the first target CT image to obtain a second contour image of the target object. Finally, terminal 101 constructs a three-dimensional volume of the target object based on each first contour image and each second contour image.

[0035] Server 103 can be a business server providing various services. It should be noted that server 103 can be either hardware or software. When server 103 is hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or as a single server. When server 103 is software, it can be implemented as multiple software programs or software modules (e.g., used to provide distributed services), or as a single software program or software module; no specific limitations are made here.

[0036] Alternatively, the system architecture may not include server 103. In other words, server 103 may be an optional device in the embodiments of this specification. That is, the method provided in the embodiments of this specification can be applied to a system structure that only includes terminal 101. The embodiments of this application do not limit this.

[0037] It should be understood that Figure 1 The number of terminals, networks, and servers shown is only illustrative; the number can be any number of terminals, networks, and servers depending on the implementation requirements.

[0038] Please see Figure 2 , Figure 2 This is a flowchart illustrating a method for constructing a 3D volume based on CT images, provided in an embodiment of this application. The execution entity in this embodiment can be a terminal executing the 3D volume construction based on CT images, a processor within the terminal executing the 3D volume construction method based on CT images, or a 3D volume construction service based on CT images within the terminal executing the 3D volume construction method based on CT images. For ease of description, the following example uses a processor within a terminal as the execution entity to illustrate the specific execution process of the 3D volume construction method based on CT images.

[0039] like Figure 2 As shown, a three-dimensional volume construction method based on CT images can include at least:

[0040] S202. Acquire CT sequence images of the target object. In response to the user's manual contour segmentation operation on the first target CT image in the CT sequence images, perform first contour segmentation on the first target CT image to obtain the first contour image of the target object.

[0041] Optionally, given that current segmentation methods often suffer from performance limitations of automatic segmentation models, making them unsuitable for complex scenarios, and that manual segmentation requires adjusting pixel thresholds for each image, resulting in a time-consuming and cumbersome process, it is crucial to consider that the accuracy of segmentation results directly impacts the accuracy of 3D reconstruction, thereby affecting the user's observation and judgment of the target object. Therefore, it is necessary to reduce user interaction time and operational steps while ensuring the accuracy of the segmentation results.

[0042] Optionally, when a user needs to observe a target object, they can first acquire the target object's computer-generated image data, such as CT sequence images. CT sequence images are X-ray computed tomography slices of the target object obtained during a single imaging process from various angles. All CT images in the sequence can be combined to create the target object's three-dimensional information. When segmenting the target object's contour based on user interaction, the user can directly observe the segmentation results of the CT image, and the target object contour determined based on the user's own operation accurately meets the user's needs. Manually segmenting all CT images would result in excessively long segmentation times and overly cumbersome operations, while manually segmenting only a small number of initial contours might lead to inaccurate contours.

[0043] Based on this, all CT images can be filtered, and manual contour segmentation can be performed on a portion of the CT images that meet the preset segmentation conditions, while automatic segmentation can be performed on the other portion. When selecting CT images for manual contour segmentation, preset segmentation conditions can be set based on factors such as the image's clarity and the angle of the target object in the image. For example, manual contour segmentation can be performed on CT images where the target object's edges are not clear, or on CT images of the target object taken from different angles. In other words, the CT images that meet the preset segmentation conditions are used as the first target CT images. In response to the user's manual contour segmentation operation on the first target CT image, the first target CT image is segmented into its first contour to obtain the first contour image of the target object.

[0044] Optionally, the first contour segmentation can employ various methods, such as using a segmentation tool to outline the target object's contour edge or manually clicking on the target object's contour edge points. Users can choose from these methods, or the computer can automatically determine the appropriate segmentation method based on the quality of the current CT image. Specifically, for first target CT images with clearly defined boundaries, the GrabCut algorithm is used; users simply need to manually circle the area to be segmented, and the segmentation tool can automatically segment the target object's contour within the circled area. For first target CT images with indistinct boundaries, an auxiliary segmentation algorithm is used, requiring users to manually click on the target object's edge points to segment its contour. The resulting first contour image is the target object's contour image obtained through user-computer interaction.

[0045] S204. Based on each first contour image, actively perform second contour segmentation on the second target CT image in the CT sequence image excluding the first target CT image to obtain the second contour image of the target object.

[0046] Optionally, as can be seen from the above embodiments, the first target CT image is a CT image that has been screened and requires manual contour segmentation. Correspondingly, the target object contours in the other CT images in the CT sequence images excluding the first target CT image, i.e. the second target CT image, can meet the automatic contour segmentation conditions. This means that the second target CT image can be actively segmented to obtain the second contour image of the target object. This directly reduces the segmentation time of the second target CT image, speeds up the segmentation of the CT sequence images, and improves the three-dimensional reconstruction efficiency of the target object. In addition, considering that the target object contours in the first contour image are accurate, the second contour segmentation of the second target CT image can be directly performed based on the first contour image to ensure that accurate contour segmentation results are obtained in the second contour image.

[0047] S206. Construct a three-dimensional object based on each first contour image and each second contour image.

[0048] Optionally, after obtaining the first contour image through first contour segmentation and the second contour image through second contour segmentation, the contour of the target object at each angle in the CT sequence image of the target object is segmented. Then, a three-dimensional volume of the target object can be constructed based on each first contour image and each second contour image. When constructing the three-dimensional volume, various existing three-dimensional reconstruction methods can be used, and this application embodiment does not limit this. The first contour image is an image obtained by the user through manual operation and interaction with the computer. Therefore, the target object contour in the first contour image is accurate and meets the user's needs. Thus, the target object contour in the second contour image obtained based on the target object contour in the first contour image is also accurate. At the same time, the actively performed second contour segmentation reduces the segmentation time of a large number of second target CT images. Therefore, this application embodiment can accelerate the construction efficiency of the three-dimensional volume of the target object and ensure the accuracy of the constructed three-dimensional volume of the target object.

[0049] This application provides a method for constructing a 3D volume based on CT images. The method involves acquiring a CT sequence image of a target object, responding to a user's manual contour segmentation operation on a first target CT image within the CT sequence image, performing a first contour segmentation on the first target CT image to obtain a first contour image of the target object, and then actively performing a second contour segmentation on a second target CT image (excluding the first target CT image) within the CT sequence image based on each first contour image to obtain a second contour image of the target object, and finally constructing a 3D volume of the target object based on each first contour image and each second contour image. Since the second contour segmentation is an active processing and calculation of the manually segmented first contour images, obtaining the second contour image avoids manual operation on a large number of CT images, reduces human-computer interaction, accelerates the 3D volume construction efficiency of the target object, and ensures the accuracy of the 3D volume construction by using manually segmented first contour images.

[0050] Please see Figure 3 , Figure 3 This is a flowchart illustrating a three-dimensional volume construction method based on CT images, provided as an embodiment of this application.

[0051] like Figure 3 As shown, a three-dimensional volume construction method based on CT images can include at least:

[0052] S302. Acquire CT sequence images of the target object. In response to the user's manual contour segmentation operation on the initial CT image in the CT sequence images, perform first contour segmentation on the initial CT image and use the first contour segmentation result as the reference contour image of the target object.

[0053] Optionally, the CT sequence images of the target object typically contain slices of the target object from multiple angles. These slices are arranged sequentially, for example, 5 slices for the left view, 5 slices for the left oblique view, 5 slices for the right view, and 5 slices for the right oblique view, resulting in a CT sequence image containing a total of 20 CT images. Usually, the contour of the target object is not significantly different at the same angle, and multiple images from the same viewpoint can be segmented based on a manually determined standard reference contour. However, the contour of the target object may differ significantly at different angles. Therefore, for CT images from different angles, it may be necessary to manually segment and determine different standard reference contours separately. In addition, images with unclear contour edges in the same viewpoint CT images may also require manual segmentation.

[0054] Furthermore, to obtain an accurate target material outline, the target object outline in the CT image at a specific angle is first determined. This requires selecting an initial CT image from the CT sequence and performing manual contour segmentation on this initial image to obtain the first contour image. When determining the initial CT image, the user can choose any image from the sequence or use the default first image; this embodiment does not limit the choice.

[0055] Optionally, after segmenting the initial CT image, the resulting first contour image is used as the first contour image. When selecting the next first target CT image, the segmentation result of the initial CT image can be used as a reference; that is, the first contour segmentation result is used as the reference contour image of the target object. Other CT images in the CT sequence are compared with the reference contour image to determine the CT image that satisfies a preset contour relationship as the next image for manual contour segmentation. The preset contour relationship can be the structural correlation, angular correlation, etc., of the target object contour between the two images. In other words, the first target CT image for first contour segmentation is the initial CT image in the CT sequence and the CT images that differ from the initial CT image in satisfying the preset contour relationship.

[0056] S304. Based on the sorting of each CT image in the CT sequence image, calculate the similarity of the contour structure of the target object between the CT image after the initial CT image in the CT sequence image and the reference contour image.

[0057] Optionally, for ease of description, please refer to Figure 4 , Figure 4 This is a schematic diagram illustrating a process for segmenting target objects in CT sequence images, provided as an embodiment of this application. Figure 4As shown, after starting the process, input CT sequence images, determine the first target CT image that needs manual contour segmentation, perform first contour segmentation on the first target CT image, and use the first contour segmentation result as the reference contour image of the target object; after obtaining the reference contour image, based on the order of each CT image in the CT sequence image, calculate the contour structure similarity of the CT images after the initial CT image in the CT sequence image with respect to the reference contour image with respect to the target object.

[0058] Optionally, contour structure similarity (SSIM) is a metric that measures the similarity between two images. A high contour structure similarity between a CT image and a reference contour image indicates that the target object's contour structure is similar in both images, making automatic segmentation potentially feasible. Conversely, a low contour structure similarity indicates that the target object's contour structure is dissimilar, requiring manual initial contour segmentation. Therefore, calculating the contour structure similarity between the initial CT image and the reference contour image with respect to the target object helps identify the next target CT image requiring manual segmentation.

[0059] S306. When a CT image with a contour structure similarity less than the reference similarity threshold is determined to be different, the calculation is paused and in response to the user's manual contour segmentation operation on the different CT image, the first contour segmentation is performed on the different CT image, the reference contour image is replaced as the latest reference contour image based on the first contour segmentation result, and the reference similarity threshold is updated based on the latest reference contour image.

[0060] Optionally, since a low similarity in contour structure between a CT image and a reference contour image indicates that the target object's contour structure is dissimilar between the two images, manual first contour segmentation is considered necessary. Therefore, when determining whether the contour structure similarity is low, a reference similarity threshold can be set, defining similarities below this threshold as low similarity. The reference similarity threshold can be set based on the contour structure of the current reference contour image, allowing it to adaptively iterate as the current reference contour image changes, thus accurately adapting to complex CT image segmentation scenarios.

[0061] Please see Figure 4When a CT image with a contour structure similarity less than the reference similarity threshold is identified as a difference, this difference is the image that needs to be segmented into its first contour. At this point, the calculation needs to be paused and the first contour segmentation should be performed in response to the user's manual contour segmentation operation on the difference CT image. After segmentation, the structural similarity needs to be calculated for subsequent CT images. At this point, it is necessary to identify CT images that are not similar to the latest first contour image and perform the next manual segmentation. Then, the reference contour image is replaced as the latest reference contour image based on the first contour segmentation result, and the reference similarity threshold is updated based on the latest reference contour image.

[0062] S308. Continue to perform calculations based on the latest reference contour image and the updated reference similarity threshold until all CT images in the CT sequence image except the initial CT image have been traversed, and the first contour image with all historical reference contour images as the target object has been obtained.

[0063] For further information, please refer to [link / reference]. Figure 4 The calculation continues based on the latest reference contour image and the updated reference similarity threshold until all CT images in the CT sequence except the initial CT image have been traversed. That is, when the last image in the CT sequence is reached, it means that the images in the CT images that need to be manually segmented have been segmented into the first contour. At this point, all historical reference contour images are used as the first contour images of the target object, thus completing the first contour segmentation process for the target object. This achieves high-quality segmentation of CT sequence images with minimal user interaction.

[0064] S3010. Based on the distance between each pixel in each first contour image and the contour edge of the target object, determine the distance matrix of each first contour image.

[0065] Optionally, after determining the first contour images in the CT sequence, the remaining unsegmented images in the CT sequence are those with high structural similarity to the two contour images preceding and following them. For example, if the first and fifth images in the CT sequence are the first contour images, then the contours of the second to fourth images are similar to the first image, and the contour change trend is between that of the first and fifth images. At this point, the distance matrix of each first contour image can be determined based on the distance between each pixel in each first contour image and the contour edge of the target object. The distance matrix is ​​calculated by setting a minimum distance from each pixel to the contour edge of the target object. If the pixel is inside the contour edge, the distance is set to negative; if the pixel is outside the contour edge, the distance is set to positive; pixels on the contour edge of the target object have a distance of zero from the contour edge. After the above operations, the distance matrix of the image can be obtained.

[0066] S3012. Two adjacent first contour images are defined as a first contour image group. Based on the distance matrix of the first contour images in each first contour image group, the second target CT image located between the first contour images in each first contour image group is actively segmented into a second contour to obtain the second contour image of the target object.

[0067] Optionally, please refer to Figure 4 At this point, a second contour segmentation can be performed on the second target CT image (excluding the first target CT image) in the CT sequence. For ease of description, two adjacent first contour images are defined as a first contour image group. Based on the distance matrix of the first contour images in each first contour image group, the second target CT image located between the first contour images in each first contour image group is actively segmented to obtain the second contour image of the target object. This reduces the user's operation time.

[0068] In a preferred embodiment, interpolation can be performed on the second target CT image located between the first contour images in each first contour image group based on the distance matrix of each first contour image group. Interpolation involves interpolating a continuous function onto discrete data, ensuring that the continuous curve passes through all given discrete data points. Interpolation can estimate the approximate value of the function at other points by considering the function's values ​​at a finite number of points, and thus can be used to fill the gaps between pixels during image transformation.

[0069] S3014. Determine the standard contour of the target object based on each first contour image and each second contour image, and construct the three-dimensional volume of the target object based on the standard contour.

[0070] Optionally, please refer to Figure 4 After segmenting the target object's contour image from CT sequence images, the standard contour of the target object can be determined first based on each first contour image and each second contour image. When determining the standard contour, the initial contour of the target object can be determined based on each first contour image and each second contour image. At this point, the initial contour may not be refined enough. Therefore, it can be further refined based on an active contour model. The active contour method, also known as the snake method, is an iterative region-growing image segmentation algorithm. Using the active contour algorithm, an initial curve can be specified on the image, and then the active contour function is used to evolve the curve towards the object boundary to determine the standard contour of the target object. Finally, the target object is reconstructed in three dimensions based on the standard contour to obtain the three-dimensional volume of the target object.

[0071] Furthermore, after obtaining the three-dimensional volume of the target object, in order to facilitate users to observe the relationship between the target three-dimensional volume and the original three-dimensional volume, the maximum intensity projection algorithm can be used to display different colors for the original three-dimensional volume to which the target object belongs and the three-dimensional volume of the target object, so as to distinguish the target three-dimensional volume from the original three-dimensional volume. The isosurface algorithm is used to display the reconstructed three-dimensional segment of the target object and the CT sequence image in the preset coordinate axis of the display interface, so as to facilitate users to observe the results and make corresponding adjustments.

[0072] In this embodiment, a method for constructing a 3D volume based on CT images is provided. The method selects a first target CT image to be manually segmented by comparing the contour structure similarity between each CT image and the previous reference contour image. First contour segmentation is performed on the first target CT image based on user-computer interaction. Then, based on the distance matrix of the first target CT image, interpolation is performed on second target CT images between adjacent groups of first target CT images to achieve second contour segmentation of the second target CT images, reducing user operation time. After obtaining accurate segmentation results, a 3D volume of the target object is constructed, and the rendered 3D volume and the CT image are displayed on the same coordinate axis for easy viewing of the segmentation status of the CT images in various directions.

[0073] Please see Figure 5 , Figure 5 This is a structural block diagram of a three-dimensional volume construction device based on CT images, provided as an embodiment of this application. Figure 5 As shown, the 3D volume construction device 500 based on CT images includes:

[0074] The first segmentation module 510 is used to acquire CT sequence images of the target object, and in response to the user's manual contour segmentation operation on the first target CT image in the CT sequence images, performs first contour segmentation on the first target CT image to obtain the first contour image of the target object.

[0075] The second segmentation module 520 is used to actively perform second contour segmentation on the second target CT image in the CT sequence image excluding the first target CT image based on each first contour image, so as to obtain the second contour image of the target object.

[0076] The 3D volume construction module 530 is used to construct the 3D volume of the target object based on each first contour image and each second contour image.

[0077] Optionally, the first target CT image is the initial CT image in the CT sequence image and the difference CT image that satisfies the preset contour relationship with the initial CT image.

[0078] Optionally, the first segmentation module 510 is further configured to, in response to a user's manual contour segmentation operation on the initial CT image in the CT sequence images, perform first contour segmentation on the initial CT image and use the first contour segmentation result as a reference contour image of the target object; based on the order of each CT image in the CT sequence images, sequentially calculate the contour structure similarity between the CT images after the initial CT image in the CT sequence images and the reference contour image with respect to the target object; when a CT image with a contour structure similarity less than the reference similarity threshold is determined to be different, pause the calculation and, in response to the user's manual contour segmentation operation on the different CT image, perform first contour segmentation on the different CT image, replace the reference contour image as the latest reference contour image according to the first contour segmentation result, and update the reference similarity threshold based on the latest reference contour image; continue to perform calculations based on the latest reference contour image and the updated reference similarity threshold until all CT images in the CT sequence images except the initial CT image have been traversed, and all historical reference contour images have been used as the first contour image of the target object.

[0079] Optionally, the second segmentation module 520 is further configured to determine the distance matrix of each first contour image based on the distance between each pixel in each first contour image and the contour edge of the target object; to define two adjacent first contour images as a first contour image group; and to actively perform second contour segmentation on the second target CT image located between the first contour images in each first contour image group according to the distance matrix of the first contour images in each first contour image group, so as to obtain the second contour image of the target object.

[0080] Optionally, the second contour segmentation is performed as an interpolation operation.

[0081] Optionally, the three-dimensional volume construction module 530 is also used to determine the standard contour of the target object based on each first contour image and each second contour image, and to construct the three-dimensional volume of the target object based on the standard contour.

[0082] Optionally, the 3D volume construction module 530 is also used to determine the initial contour of the target object based on each first contour image and each second contour image; and to standardize the initial contour based on the active contour model to determine the standard contour of the target object.

[0083] In this embodiment, a 3D volume construction device based on CT images is provided. A first segmentation module is used to acquire CT sequence images of a target object. In response to a user's manual contour segmentation operation on a first target CT image within the CT sequence images, the first target CT image is segmented to obtain a first contour image of the target object. A second segmentation module is used to actively segment a second target CT image (excluding the first target CT image) within the CT sequence images based on each first contour image to obtain a second contour image of the target object. A 3D volume construction module is used to construct a 3D volume of the target object based on each first contour image and each second contour image. Since the second contour segmentation is an active processing calculation of the manually segmented first contour images, obtaining the second contour image avoids manual operation on a large number of CT images, reduces human-computer interaction, accelerates the 3D volume construction efficiency of the target object, and ensures the accuracy of the 3D volume construction of the target object from the manually segmented first contour images.

[0084] This application also provides a computer storage medium that can store multiple instructions adapted for loading by a processor and executing the steps of any of the methods described in the above embodiments.

[0085] Please see Figure 6 , Figure 6 This is a schematic diagram of the structure of a terminal provided in an embodiment of this application. Figure 6 As shown, terminal 600 may include: at least one terminal processor 601, at least one network interface 604, user interface 603, memory 605, and at least one communication bus 602.

[0086] The communication bus 602 is used to enable communication between these components.

[0087] The user interface 603 may include a display screen and a camera. Optionally, the user interface 603 may also include a standard wired interface and a wireless interface.

[0088] The network interface 604 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0089] The terminal processor 601 may include one or more processing cores. The terminal processor 601 connects to various parts within the terminal 600 using various interfaces and lines, and performs various functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 605, and by calling data stored in the memory 605. Optionally, the terminal processor 601 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The terminal processor 601 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content to be displayed on the screen; and the modem handles wireless communication. It is understood that the modem may also not be integrated into the terminal processor 601 and may be implemented as a separate chip.

[0090] The memory 605 may include random access memory (RAM) or read-only memory (ROM). Optionally, the memory 605 may include a non-transitory computer-readable storage medium. The memory 605 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 605 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory 605 may also be at least one storage device located remotely from the aforementioned terminal processor 601. Figure 6 As shown, the memory 605, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and a three-dimensional volume construction program based on CT images.

[0091] exist Figure 6In the terminal 600 shown, the user interface 603 is mainly used to provide an input interface for the user and to acquire user input data; while the terminal processor 601 can be used to call the 3D volume construction program based on CT images stored in the memory 605, and specifically perform the following operations:

[0092] Acquire CT sequence images of the target object; in response to the user's manual contour segmentation operation on the first target CT image in the CT sequence images, perform first contour segmentation on the first target CT image to obtain the first contour image of the target object.

[0093] Based on each first contour image, the second contour segmentation is actively performed on the second target CT image in the CT sequence image excluding the first target CT image to obtain the second contour image of the target object;

[0094] Construct a three-dimensional object based on each first contour image and each second contour image.

[0095] In some embodiments, the first target CT image is the initial CT image in the CT sequence image and the difference CT image that satisfies a preset contour relationship with the initial CT image.

[0096] In some embodiments, when the terminal processor 601 performs a first contour segmentation on the first target CT image in response to a user's manual contour segmentation operation on the first target CT image in the CT sequence images to obtain a first contour image of the target object, it specifically performs the following steps: in response to a user's manual contour segmentation operation on the initial CT image in the CT sequence images, it performs a first contour segmentation on the initial CT image and uses the first contour segmentation result as a reference contour image of the target object; based on the order of each CT image in the CT sequence images, it sequentially calculates the contour structure similarity of the CT images after the initial CT image in the CT sequence images with respect to the reference contour image with respect to the target object; when it is determined that there is a CT image with a different contour structure similarity less than the reference similarity threshold, it pauses the calculation and performs a first contour segmentation on the CT image in response to a user's manual contour segmentation operation on the different CT image, replaces the reference contour image with the latest reference contour image according to the first contour segmentation result, and updates the reference similarity threshold based on the latest reference contour image; it continues to perform the calculation based on the latest reference contour image and the updated reference similarity threshold until all CT images in the CT sequence images except the initial CT image have been traversed, and all historical reference contour images have been used as the first contour image of the target object.

[0097] In some embodiments, when the terminal processor 601 performs second contour segmentation on a second target CT image (excluding the first target CT image) in a CT sequence image based on each first contour image to obtain a second contour image of the target object, it specifically performs the following steps: determining a distance matrix for each first contour image based on the distance between each pixel in each first contour image and the contour edge of the target object; determining two adjacent first contour images as a first contour image group; and actively performing second contour segmentation on the second target CT image located between the first contour images in each first contour image group according to the distance matrix of the first contour images in each first contour image group to obtain a second contour image of the target object.

[0098] In some embodiments, the second contour segmentation is an interpolation operation.

[0099] In some embodiments, when the terminal processor 601 constructs a three-dimensional volume of the target object based on each first contour image and each second contour image, it specifically performs the following steps: determining the standard contour of the target object based on each first contour image and each second contour image, and constructing a three-dimensional volume of the target object based on the standard contour.

[0100] In some embodiments, when the terminal processor 601 executes the following steps when determining the standard contour of the target object based on each first contour image and each second contour image: determining the initial contour of the target object based on each first contour image and each second contour image; and standardizing the initial contour based on the active contour model to determine the standard contour of the target object.

[0101] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0102] The modules described as separate components may or may not be physically separate. Similarly, the components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0103] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When these computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this specification are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in or transmitted through a computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The aforementioned available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., Digital Versatile Discs (DVDs)), or semiconductor media (e.g., Solid State Disks (SSDs)).

[0104] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0105] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0106] The above is a description of a method, apparatus, storage medium, and terminal for constructing three-dimensional volumes based on CT images provided in this application. For those skilled in the art, based on the ideas of the embodiments of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for constructing a three-dimensional volume based on CT images, characterized in that, The method includes: The system acquires CT sequence images of the target object. In response to a user's manual contour segmentation operation on a first target CT image in the CT sequence images, it performs first contour segmentation on the first target CT image to obtain a first contour image of the target object. The first target CT image is an initial CT image in the CT sequence images and a difference CT image that satisfies a preset contour relationship with the initial CT image. Based on each first contour image, the second contour segmentation is actively performed on the second target CT image in the CT sequence image excluding the first target CT image to obtain the second contour image of the target object; Construct the three-dimensional volume of the target object based on each first contour image and each second contour image; The step of actively segmenting the second target CT image (excluding the first target CT image) in the CT sequence image based on each first contour image to obtain the second contour image of the target object includes: Based on the distance between each pixel in each first contour image and the contour edge of the target object, a distance matrix for each first contour image is determined; two adjacent first contour images are defined as a first contour image group; according to the distance matrix of the first contour images in each first contour image group, the second target CT image located between the first contour images in each first contour image group is actively segmented into a second contour to obtain the second contour image of the target object. The step of responding to a user's manual contour segmentation operation on a first target CT image in the CT sequence images, performing a first contour segmentation on the first target CT image to obtain a first contour image of the target object, includes: In response to a user's manual contour segmentation operation on the initial CT image in the CT sequence images, the initial CT image is subjected to first contour segmentation, and the first contour segmentation result is used as a reference contour image of the target object. Based on the order of each CT image in the CT sequence, the similarity of the contour structure of the target object between the CT images after the initial CT image in the CT sequence and the reference contour image is calculated sequentially. When a CT image with a contour structure similarity less than a reference similarity threshold is determined to be different, the calculation is paused and in response to the user's manual contour segmentation operation on the different CT image, a first contour segmentation is performed on the different CT image, the reference contour image is replaced as the latest reference contour image based on the first contour segmentation result, and the reference similarity threshold is updated based on the latest reference contour image. The calculation continues based on the latest reference contour image and the updated reference similarity threshold until all CT images in the CT sequence image except the initial CT image have been traversed, and all historical reference contour images are used as the first contour image of the target object.

2. The method according to claim 1, characterized in that, The second contour segmentation is an interpolation operation.

3. The method according to claim 1, characterized in that, The step of constructing the three-dimensional volume of the target object based on each first contour image and each second contour image includes: The standard contour of the target object is determined based on each first contour image and each second contour image, and the three-dimensional volume of the target object is constructed based on the standard contour.

4. The method according to claim 3, characterized in that, The determination of the standard contour of the target object based on each first contour image and each second contour image includes: The initial contour of the target object is determined based on each first contour image and each second contour image; The initial contour is standardized based on the active contour model to determine the standard contour of the target object.

5. A three-dimensional volume construction device based on CT images, characterized in that, The device includes: The first segmentation module is used to acquire CT sequence images of the target object, and in response to a user's manual contour segmentation operation on a first target CT image in the CT sequence images, performs first contour segmentation on the first target CT image to obtain a first contour image of the target object; the first target CT image is an initial CT image in the CT sequence images and a difference CT image that satisfies a preset contour relationship with the initial CT image; The second segmentation module is used to actively perform second contour segmentation on the second target CT image (excluding the first target CT image) in the CT sequence image based on each first contour image, so as to obtain the second contour image of the target object. A 3D volume construction module is used to construct a 3D volume of the target object based on each first contour image and each second contour image. The second segmentation module is further configured to determine the distance matrix of each first contour image based on the distance between each pixel in each first contour image and the contour edge of the target object; to define two adjacent first contour images as a first contour image group; and to actively perform second contour segmentation on the second target CT image located between the first contour images in each first contour image group according to the distance matrix of the first contour images in each first contour image group, so as to obtain the second contour image of the target object. The first segmentation module is configured to, in response to a user's manual contour segmentation operation on the initial CT image in the CT sequence images, perform a first contour segmentation on the initial CT image and use the first contour segmentation result as a reference contour image of the target object; based on the order of each CT image in the CT sequence images, sequentially calculate the contour structure similarity between the CT images after the initial CT image in the CT sequence images and the reference contour image with respect to the target object; when a CT image with a contour structure similarity less than a reference similarity threshold is determined to be a different CT image, the calculation is paused and, in response to the user's manual contour segmentation operation on the different CT image, the first contour segmentation is performed on the different CT image, the reference contour image is replaced as the latest reference contour image according to the first contour segmentation result, and the reference similarity threshold is updated based on the latest reference contour image; the calculation continues based on the latest reference contour image and the updated reference similarity threshold until all CT images in the CT sequence images except the initial CT image have been traversed, and all historical reference contour images are used as the first contour image of the target object.

6. A computer storage medium, characterized in that, The computer storage medium stores a plurality of instructions adapted for loading by a processor and executing the steps of the method as described in any one of claims 1 to 4.

7. A terminal, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the method as described in any one of claims 1 to 4.

Images