Image generation method and device, equipment and medium
By constructing a target grid and processing data blocks, the problem of slow generation speed of maximum density projection images is solved, achieving efficient image generation and meeting the needs of real-time clinical diagnosis.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-17
- Publication Date
- 2026-03-13
AI Technical Summary
Existing methods for generating maximum density projection images involve large computational demands, resulting in slow generation speeds and making it difficult to meet the needs of real-time clinical diagnosis.
By constructing a target mesh, the 3D image data is divided into multiple data blocks. The maximum density value of each data block is calculated in advance, the maximum density value of the boundary slices is identified, the global maximum density value is determined, and the maximum density projection image is generated.
It greatly reduces the amount of computation, significantly improves the image generation speed, increases efficiency, and can quickly provide clear images for clinical diagnosis.
Smart Images
Figure CN121661283A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of image processing and is applied to online processing business scenarios such as technology and healthcare. In particular, it relates to an image generation method, apparatus, device, and medium. Background Technology
[0002] In the field of medical image processing, especially in the area of 3D reconstruction, the 3D volumetric data acquired by imaging technologies such as CT, MRI, and PET contains rich anatomical and physiological information. To facilitate clinical observation and analysis, it is often necessary to compress the 3D volumetric data into 2D images, leading to the development of volume rendering techniques such as maximum density projection (MIP), average density projection, minimum density projection, and summation projection.
[0003] Among these technologies, maximum density projection (MIP) is a key technique. Its principle is as follows: for each pixel in the final 2D image, a line of sight (projection ray) perpendicular to the projection plane is emitted from that point, penetrating the entire 3D volumetric data. MIP then finds the highest value among all voxels (3D pixels) along this ray and assigns it to the corresponding pixel in the 2D image. In this way, MIP can highlight the structures with the highest density in 3D data, making it valuable for applications such as vascular imaging and bone examination.
[0004] However, in existing technologies, the generation speed of maximum density projection images is relatively slow. This is because traditional methods, when processing three-dimensional volume data, require comparing each voxel through which the projection ray passes to find the maximum value. This process involves a large amount of computation, especially when processing high-resolution three-dimensional image data, where the computational load increases exponentially, resulting in excessively long generation times. This makes it difficult to meet the needs of real-time clinical diagnosis and hinders the further promotion and application of MIP technology. Summary of the Invention
[0005] The purpose of this application is to provide an image generation method, apparatus, computer device, and storage medium to solve the problem that existing maximum density projection image generation methods are slow and inefficient due to excessive computation.
[0006] Firstly, an image generation method is provided, which adopts the following technical solution: Acquire 3D image data and construct a target mesh based on a preset cutting direction. Divide the 3D image data into multiple data blocks based on the target mesh, and calculate the maximum density value of all voxels in each data block. Determine the maximum density value of all complete data blocks within the target volume of the 3D image data based on the maximum density values of all voxels. Identify boundary slices of the target volume along the projection direction and obtain the maximum density value of the boundary slices. Determine the global maximum density value of the target volume based on the maximum density values of the complete data blocks and the boundary slices. Generate a maximum density projection image of the target volume based on the global maximum density value, the maximum density value of the complete data blocks, and the maximum density value of the boundary slices.
[0007] Secondly, an image generation apparatus is provided, which adopts the following technical solution: The construction module is used to acquire 3D image data and construct a target mesh based on the preset cutting direction for the 3D image data; The partitioning module is used to divide the 3D image data into multiple data blocks based on the target grid and calculate the maximum density value of all voxels in each data block; The first determining module is used to determine the maximum density value of all complete data blocks within the target volume of the three-dimensional image data based on the maximum density value of all voxels. The first acquisition module is used to identify the boundary slices of the target volume in the projection direction and obtain the maximum density value of the boundary slices; The second determining module is used to determine the global maximum density value of the target volume based on the maximum density value of the complete data block and the maximum density value of the boundary slice; The generation module is used to generate a maximum density projection image of the target volume based on the global maximum density value, the maximum density value of the complete data block, and the maximum density value of the boundary slice.
[0008] Thirdly, a computer device is provided, which adopts the following technical solution: Acquire 3D image data and construct a target mesh based on a preset cutting direction. Divide the 3D image data into multiple data blocks based on the target mesh, and calculate the maximum density value of all voxels in each data block. Determine the maximum density value of all complete data blocks within the target volume of the 3D image data based on the maximum density values of all voxels. Identify boundary slices of the target volume along the projection direction and obtain the maximum density value of the boundary slices. Determine the global maximum density value of the target volume based on the maximum density values of the complete data blocks and the boundary slices. Generate a maximum density projection image of the target volume based on the global maximum density value, the maximum density value of the complete data blocks, and the maximum density value of the boundary slices.
[0009] Fourthly, a computer-readable storage medium is provided, which adopts the following technical solution: Acquire 3D image data and construct a target mesh based on a preset cutting direction. Divide the 3D image data into multiple data blocks based on the target mesh, and calculate the maximum density value of all voxels in each data block. Determine the maximum density value of all complete data blocks within the target volume of the 3D image data based on the maximum density values of all voxels. Identify boundary slices of the target volume along the projection direction and obtain the maximum density value of the boundary slices. Determine the global maximum density value of the target volume based on the maximum density values of the complete data blocks and the boundary slices. Generate a maximum density projection image of the target volume based on the global maximum density value, the maximum density value of the complete data blocks, and the maximum density value of the boundary slices.
[0010] Compared with existing technologies, the embodiments of this application have the following main advantages: By constructing a target mesh, the 3D image data is accurately divided into multiple data blocks, and the maximum density value of voxels within each data block is calculated in advance. During processing, it is not necessary to compare each voxel that the projection ray passes through one by one; instead, the operation is based on the maximum density value of the data block. First, the maximum density value of the complete data block within the target volume is determined, then the maximum density value of the boundary slice is identified, and finally, the global maximum density value is determined. This block-based processing method greatly reduces the amount of computation and avoids the problem of exponential growth in computational load caused by comparing voxels one by one when processing high-resolution 3D image data. Finally, a maximum density projection image is generated based on these pre-calculated and processed maximum density values, significantly improving the image generation speed and effectively increasing efficiency. Attached Figure Description
[0011] To more clearly illustrate the solutions in this application, the accompanying drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0012] Figure 1 This is an exemplary system architecture diagram to which this application can be applied; Figure 2 A flowchart of an embodiment of the image generation method according to this application; Figure 3 This is a schematic diagram of the structure of an embodiment of the image generation apparatus according to this application; Figure 4 This is a schematic diagram of the structure of one embodiment of the computer device according to this application. Detailed Implementation
[0013] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein in the specification of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and foregoing drawings of this application, are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or foregoing drawings of this application are used to distinguish different objects, not to describe a particular order.
[0014] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.
[0015] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.
[0016] like Figure 1 As shown, system architecture 100 may include terminal device 101, network 102, and server 103. Terminal device 101 may be a laptop 1011, tablet 1012, or mobile phone 1013. Network 102 is used as a medium to provide a communication link between terminal device 101 and server 103. Network 102 may include various connection types, such as wired, wireless communication links, or fiber optic cables.
[0017] Users can use terminal device 101 to interact with server 103 via network 102 to receive or send messages, etc. Various communication client applications can be installed on terminal device 101, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social media platform software, etc.
[0018] Terminal device 101 can be various electronic devices with a display screen and support web browsing. In addition to laptops 1011, tablets 1012, or mobile phones 1013, terminal device 101 can also be an e-book reader, an MP3 player (Moving Picture Experts Group Audio Layer III), an MP4 player (Moving Picture Experts Group Audio Layer IV), a laptop computer, and a desktop computer, etc.
[0019] Server 103 can be a server that provides various services, such as a backend server that provides support for the pages displayed on terminal device 101.
[0020] It should be noted that the image generation method provided in this application embodiment is generally executed by a server / terminal device, and correspondingly, the image generation device is generally located in the server / terminal device.
[0021] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.
[0022] Continue to refer to Figure 2 A flowchart of an embodiment of the image generation method according to this application is shown. The image generation method includes the following steps: Step S201: Acquire 3D image data and construct a target mesh based on the preset cutting direction for the 3D image data; Three-dimensional image data is acquired through medical imaging technologies such as CT, MRI, and PET, and is a collection of information used to characterize the internal structure of the human body or object in three-dimensional space. For example, in a CT scan, different tissues absorb X-rays to different degrees, resulting in data points with different gray values, which together constitute the three-dimensional image data.
[0023] The cutting direction includes, but is not limited to, the horizontal plane, the coronal plane, and the axial plane, which provide a standardized spatial reference framework for medical image processing. For example, the horizontal plane divides the object corresponding to the 3D image data into upper and lower parts, the coronal plane divides the object into front and back parts, and the axial plane divides the object into left and right parts. When constructing the target mesh and performing projection operations, the cutting direction can ensure that the processing conforms to the actual physiological structure of the object.
[0024] The target mesh is a spatial partitioning structure built based on a preset cutting direction for 3D image data. For example, an adaptive mesh can dynamically adjust according to the characteristics and requirements of the 3D image data, dividing the 3D space into multiple regular or irregular small regions. For instance, when dividing 3D image data into multiple data blocks, the target mesh can serve as a partitioning basis, making the data block division more reasonable, improving the efficiency and accuracy of data processing, and adapting to 3D image data of different shapes and features.
[0025] Step S202: Based on the target grid, the 3D image data is divided into multiple data blocks, and the maximum density value of all voxels in each data block is calculated. Multiple data blocks are sub-data sets formed by dividing 3D image data based on a target grid. Each data block contains information about the 3D image data within a specific grid region. For example, a head CT 3D image may be divided into multiple small data blocks according to a target grid, with each data block corresponding to a local region of the head.
[0026] Voxel, short for three-dimensional pixel, is the basic unit in three-dimensional image data, similar to a pixel in a two-dimensional image. Each voxel represents a tiny volume region in three-dimensional space and has a corresponding value, which usually represents information such as density, signal strength, or tracer concentration.
[0027] The maximum density value refers to the highest numerical value among a group of voxels, representing the highest level of density, signal strength, or tracer concentration in that group of voxels. For example, when calculating the maximum density value of all voxels in each data block, the largest value is found among the many voxels contained in that data block.
[0028] Step S203: Based on the maximum density value of all voxels, determine the maximum density value of all complete data blocks within the target volume of the 3D image data. The target volume refers to the integrated portion currently undergoing Maximum Intensity Projection (MIP) processing; it is a specific region within the 3D image data. For example, when performing MIP processing on the lungs, the target volume is the 3D image data region of the lungs. Clearly defining the target volume allows for focused processing, enabling MIP operations to be performed only on the region of interest, reducing unnecessary calculations and improving processing efficiency and targeting.
[0029] A complete data block refers to a data block that is entirely located within the target volume, meaning that all voxels of the data block are within the target volume. For example, within a rectangular target volume, a data block that is completely contained within that rectangle is a complete data block.
[0030] Step S204: Identify the boundary slices of the target volume in the projection direction and obtain the maximum density value of the boundary slices; The projection direction refers to the direction of the line of sight from a pixel in the final 2D image, perpendicular to the projection plane. In maximum density projection, the projection ray penetrates the entire 3D volume data along this direction. For example, when generating a frontal maximum density projection image, the projection direction is the direction perpendicular to the 3D volume data from the front of the image.
[0031] Boundary slices refer to the edge slices of the target volume along the projection direction, such as upper boundary slices and lower boundary slices. They are the boundaries between the target volume and the external region. For example, in the vertical projection direction, the uppermost and lowermost slices of the target volume are boundary slices.
[0032] Step S205: Determine the global maximum density value of the target volume based on the maximum density value of the complete data block and the maximum density value of the boundary slice; The global maximum density value is the maximum density value within the target volume determined based on the maximum density values of the complete data blocks and the boundary slices. For example, the maximum value among the maximum density values of all complete data blocks is first found, and then compared with the maximum density values of the boundary slices to obtain a maximum density value as the global maximum density value.
[0033] Step S206: Generate a maximum density projection image of the target volume based on the global maximum density value, the maximum density value of the complete data block, and the maximum density value of the boundary slice.
[0034] Maximum density projection (MDP) images represent a specific projection format resulting from compressing three-dimensional volumetric data into a two-dimensional image. This type of image is used to highlight the highest-density structures within the three-dimensional data, facilitating clinical observation and analysis, and has significant application value in areas such as vascular imaging and bone examination. For example, in cerebral vascular imaging, MDP images can clearly display high-density regions within blood vessels, such as narrowed areas or calcified regions, helping doctors accurately assess vascular lesions.
[0035] Specifically, after determining the global maximum density value, the maximum density value of the complete data block, and the maximum density value of the boundary slice, the projection points corresponding to the global maximum density value, the maximum density values of all complete data blocks, and the maximum density values of the boundary slice can be obtained. Then, the image is drawn based on these projection points to obtain the maximum density projection image of the target volume.
[0036] In one embodiment, for three-dimensional medical image data, corresponding target meshes are constructed in three anatomical directions: horizontal, coronal, and axial. Based on the constructed target meshes, the three-dimensional data is divided into multiple discrete data blocks (hereinafter referred to as "recording blocks"). During the partitioning process, the number of slices in each recording block in the corresponding cutting direction, i.e., the anatomical direction, is determined. For each partitioned recording block, the maximum density of all voxels within the recording block is calculated. Further, the maximum density value of all complete recording blocks within the target volume (i.e., the integrated portion currently undergoing maximum density projection (MIP) processing) is calculated. The calculated maximum density value within the recording block and the maximum density value of the complete recording blocks within the target volume are stored in memory. If the data partitioning step fails to generate valid recording blocks (e.g., due to data size or parameter settings preventing partitioning), the target volume is integrated using a conventional maximum density projection (MIP) algorithm. This method requires traversing the data layer by layer and determining the minimum density value through global comparison. Otherwise, the following maximum density projection integration step based on recording blocks is performed. The upper and lower boundary slices of the target volume in the projection direction are determined. The matrices corresponding to the upper and lower boundary slices are analyzed to determine their positional information within the recording block partitioning. Based on this positional information, recording blocks containing the upper boundary slice and recording blocks containing the lower boundary slice are obtained. The maximum voxel density contained in these two recording blocks within the target volume integration range is extracted. The maximum density value of the complete recording block within the target volume is compared with the maximum density values of the upper and lower boundary slices to obtain the global maximum density value. When generating the final maximum density projection (MIP) image, the projection points corresponding to the global maximum density value can be recorded in the Region of Interest (ROI), the projection points corresponding to the maximum density values of the upper and lower boundary slices can be recorded in the ROI, and the projection points corresponding to the maximum density values contributed by all recording blocks within the target volume can also be recorded in the ROI. Then, it can be detected whether there are edge data regions between the upper and lower boundary slices that are not covered by the complete recording block. If not, the global maximum density value is used as the global maximum density value of the target volume; if it exists, then: a) Extract the maximum density value of the edge data region corresponding to the upper boundary slice; b) Extract the maximum density value of the edge data region corresponding to the lower boundary slice; c) Compare the maximum density value calculated in steps a) and b) with the global maximum density value obtained above, and determine the maximum density value with the largest value after comparison as the target global maximum density value of the target volume. d) Add the projection points corresponding to the maximum density values of the edge data regions corresponding to the upper boundary slice, the projection points corresponding to the maximum density values of the edge data regions corresponding to the lower boundary slice, and the projection points corresponding to the target global maximum density value to the ROI to optimize the maximum density projection image generated above.
[0037] This application constructs a target mesh to precisely divide 3D image data into multiple data blocks and pre-calculates the maximum density value of voxels within each data block. During processing, it eliminates the need to compare each voxel along the projection ray; instead, operations are performed based on the maximum density value of each data block. First, the maximum density value of the complete data block within the target volume is determined, then the maximum density value of the boundary slices is identified, and finally, the global maximum density value is determined. This block-based processing method significantly reduces computational load, avoiding the exponential growth in computational complexity caused by comparing voxels one by one in high-resolution 3D image data processing. Finally, a maximum density projection image is generated based on these pre-calculated and processed maximum density values, significantly improving image generation speed and efficiency.
[0038] In some optional implementations of this embodiment, the boundary slice includes an upper boundary slice and a lower boundary slice. Step S204 involves identifying the boundary slices of the target volume in the projection direction and obtaining the maximum density value of the boundary slices, specifically including the following steps: Identify the upper and lower boundary slices of the target volume in the projection direction; analyze the matrices corresponding to the upper and lower boundary slices to obtain the position information of the matrices in the data block partitioning; based on the position information, extract the maximum density value of the upper and lower boundary slices.
[0039] The upper boundary slice represents the two-dimensional data slice corresponding to the topmost layer of the target volume under a preset projection direction. For example, when processing three-dimensional CT images of the human head, if the projection direction is from top to bottom, then the upper boundary slice is the two-dimensional data slice corresponding to the topmost layer of the CT image of the head.
[0040] The lower boundary slice represents the two-dimensional data slice corresponding to the lowest level of the target volume under a preset projection direction. For example, in the above-mentioned processing of three-dimensional CT image data of the human head, if the projection direction is from top to bottom, the lower boundary slice is the two-dimensional data slice corresponding to the lowest layer of the CT image of the head.
[0041] The matrix is used to accurately describe the density distribution characteristics of the boundary slices. By analyzing the matrix, the position information of the boundary slices in the data block division can be obtained, and then the maximum density values of the upper and lower boundary slices can be obtained.
[0042] Location information refers to the specific position of the matrix within the data block partitioning. It represents the coordinates, range, and other positional characteristics of the matrix corresponding to the boundary slice within the overall data block partitioning system.
[0043] In one example, using human lung CT 3D image data, the process of identifying boundary slices, obtaining location information, and determining the maximum density value of corresponding data blocks is explained in detail. First, human lung CT 3D image data is acquired, and an adaptive target mesh is constructed based on a preset cutting direction, dividing the 3D image data into multiple data blocks. The upper and lower boundary slices of the target volume are identified along the projection direction (assuming it's from top to bottom). By traversing each layer of slices in the projection direction of the 3D image data, the slices intersecting the target volume at the top and bottom are determined; these are the upper and lower boundary slices. Next, the matrices corresponding to the upper and lower boundary slices are analyzed. This matrix consists of the density values of each voxel on the slice, reflecting the density distribution of the slice. The positional information of the matrix within the data block division is obtained through a matrix analysis algorithm. Based on the positional information, the maximum density values of the upper and lower boundary slices are obtained. This method avoids comparing all voxels of the entire 3D volumetric image data one by one, reducing computational load and increasing the generation speed of the maximum density projection image, enabling faster provision of the required images for clinical diagnosis.
[0044] In some optional implementations, step S206, generating a maximum density projection image of the target volume based on the global maximum density value, the maximum density value of the complete data block, and the maximum density value of the boundary slice, specifically includes the following steps: The voxel positions corresponding to the global maximum density value, the maximum density value of the complete data block, and the maximum density value of the boundary slice are obtained respectively; based on the voxel positions, the maximum density projection image of the target volume is generated.
[0045] In this context, voxel location refers to the specific spatial coordinates of a voxel, which is a three-dimensional pixel, within a pre-defined three-dimensional coordinate system in three-dimensional medical image data. For example, when processing CT three-dimensional image data of the human brain, each voxel of the brain tissue has its specific voxel location, which allows us to clearly identify the specific region of the brain in which the voxel is located.
[0046] In one example, using 3D CT images of the human brain, the process of acquiring voxel locations and generating a maximum density projection image is illustrated. First, a target mesh is constructed according to a preset cutting direction, dividing the brain CT 3D image data into multiple data blocks. The maximum density value of all voxels in each data block is calculated, thereby determining the maximum density values of complete data blocks and boundary slices within the target volume, obtaining the global maximum density value. Then, the voxel locations corresponding to the global maximum density value, the maximum density value of the complete data block, and the maximum density value of the boundary slice are obtained. For example, in a brain CT image, the algorithm locates a voxel with a global maximum density value in a specific blood vessel region of the brain; the voxel corresponding to the maximum density value of the complete data block is in the gray matter region of the brain; and the voxel corresponding to the maximum density value of the boundary slice is in the white matter region of the brain. Based on these voxel locations, a projection algorithm is used to map the voxel locations in 3D space onto a 2D projection plane. During the projection process, the pixel coordinates of the voxels in the 2D image are determined according to their locations, and the maximum density value is assigned to the corresponding pixel. By accurately acquiring voxel locations, the traditional method of calculating each voxel individually is avoided, significantly reducing the computational load and increasing the generation speed of maximum density projection images. This enables the rapid provision of clear projection images of high-density structures such as cerebral blood vessels for clinical use.
[0047] In some optional implementations, after dividing the 3D image data into multiple data blocks based on the target mesh and calculating the maximum density value of all voxels in each data block in step S202, the following steps are also included: Obtain the storage hierarchy information of the multi-level storage system; based on the storage hierarchy information, store the maximum density value of each data block in the target storage hierarchy of the hierarchical storage system.
[0048] Among them, a multi-level storage system represents a storage architecture composed of various storage devices with different speeds, capacities, and costs arranged in a hierarchical manner. It is used for the efficient management of 3D image data and the various types of data generated during its processing.
[0049] Among them, the storage tier information comes from the configuration and characteristics of the multi-level storage system and is used to characterize the attribute parameters of different storage tiers in the multi-level storage system.
[0050] The target storage tier is derived from the evaluation and selection of each storage tier in a multi-level storage system. It is a specific storage tier determined based on the characteristics of the data, the frequency of use, and the performance requirements of the storage system.
[0051] In one example, taking CT 3D images of a human liver as an example, the process of generating a maximum density projection image in medical image processing involves several steps. First, the storage hierarchy information of the multi-level storage system is acquired. Through the system's built-in storage management module, it is determined that the multi-level storage system includes a cache layer, a solid-state drive (SSD) layer, and a hard disk drive (HDD) layer, and key parameters such as storage capacity and read / write speed for each layer are identified. Next, based on the acquired storage hierarchy information, the maximum density value of each data block is stored in the target storage hierarchy of the hierarchical storage system. For the maximum density values of data blocks that are participating in real-time computation and are frequently used, they can be stored in the cache layer to ensure fast access. For the maximum density values of data blocks that have completed partial computation but may still be used later with lower frequency, they can be stored in the SSD layer. For the maximum density values of data blocks that have been processed but are not frequently used, they can be stored in the HDD layer. This storage method reduces data retrieval time, significantly improving computational efficiency and shortening image generation time when generating the maximum density projection image based on the global maximum density value.
[0052] In some optional implementations, step S205, determining the global maximum density value of the target volume based on the maximum density value of the complete data block and the maximum density value of the boundary slice, specifically includes the following steps: Obtain the maximum density value of the complete data block from the multi-level storage system; compare the maximum density value of the complete data block with the maximum density value of the boundary slice; and determine the global maximum density value of the target volume based on the comparison results.
[0053] In one example, when processing 3D CT image data of the target object's brain to generate a maximum density projection image, a target mesh is first constructed according to a preset cutting direction, dividing the 3D image data into multiple data blocks. The maximum density value of all voxels in each data block is calculated and stored in a multi-level storage system. During the stage of determining the global maximum density value of the target volume, the maximum density value of the complete data block within the target volume, as well as the maximum density value of the boundary slices, can be obtained from the multi-level storage system. Then, these maximum density values are compared, and the maximum density value with the largest maximum density value is determined as the global maximum density value of the target volume. Subsequently, based on this global maximum density value and the maximum density values of the complete data block and boundary slices, the maximum density projection image of the target volume can be generated. This method utilizes a multi-level storage system to quickly acquire data, reducing data retrieval time, improving the efficiency of determining the global maximum density value, and thus accelerating the generation speed of the maximum density projection image.
[0054] In some optional implementations, the boundary slices include upper boundary slices and lower boundary slices. After generating the maximum density projection image of the target volume based on the global maximum density value, the maximum density value of the complete data block, and the maximum density value of the boundary slices in step S206, the following steps are further included: The algorithm detects whether there are edge data regions between the upper and lower boundary slices that are not covered by complete data blocks. If so, it obtains the maximum density value of the edge data region of the upper and lower boundary slices. The maximum density values of the edge data regions of the upper and lower boundary slices are compared with the global maximum density value. Based on the comparison results, the global maximum density value is corrected to obtain the target global maximum density value of the target volume. The maximum density projection image is optimized based on the maximum density values of the edge data regions of the upper and lower boundary slices and the target global maximum density value.
[0055] Edge data regions refer to data located at the edges of boundary data blocks but within the target volume. These data may not have been fully considered during the initial data block division and processing. For example, within a boundary data block, some voxels may be located within the target volume but, due to the data block division method, are not included in the regular processing of the complete data block or boundary slice. Detecting edge data regions and obtaining their maximum density values ensures that the maximum density values of all voxels within the target volume are taken into account, thereby more accurately determining the global maximum density value of the target volume and improving the accuracy of the maximum density projection image.
[0056] The target global maximum density value refers to the highest value among all voxels within the target volume. It represents the highest level of density, signal strength, or tracer concentration within the target volume. For example, by comprehensively considering the maximum density values of the complete data block, the boundary slices, and the edge data regions, the final maximum density value within the target volume is determined as the target global maximum density value. This value is a key parameter for generating the maximum density projection image, determining the value of the corresponding pixel in the final two-dimensional image, thereby highlighting the structure with the highest density within the target volume.
[0057] In one example, in a real-world medical image processing scenario, after acquiring the patient's 3D image data, a target mesh is constructed and data blocks are divided according to the aforementioned steps. The global maximum density value of the target volume is calculated, and a maximum density projection image is generated. During further processing, detection reveals edge data regions not fully covered by data blocks between the upper and lower boundary slices. For example, when processing 3D brain CT image data, edge data regions are formed at the brain's edges due to data division and cutting directions. In this case, the system automatically obtains the maximum density value of the edge data region of the upper boundary slice, setting it to 80 HU, and the maximum density value of the edge data region of the lower boundary slice, setting it to 75 HU. These two values are compared with the previously obtained global maximum density value of 90 HU. The comparison shows that the maximum density value of the edge data region is less than the global maximum density value, so the previously obtained global maximum density value of 90 HU is still used as the target global maximum density value for the target volume. Finally, the projection points corresponding to the maximum density value of 90HU in the edge data region of the upper boundary slice, the maximum density value of 75HU in the edge data region of the lower boundary slice, and the target global maximum density value of 90HU are supplemented and recorded in the global maximum density projection image. The maximum density projection image is then optimized to make the generated image edges clearer and the information more complete, effectively improving the image quality.
[0058] In some optional implementations, after the step "detecting whether there are edge data regions between the upper and lower boundary slices that are not covered by the complete data block", the following steps are also included: If there are no edge data regions in the boundary slice that are not covered by the complete data block, then the global maximum density value is determined as the target global maximum density value of the target volume.
[0059] In one example, following the previous one, in a real medical image processing scenario, after acquiring the patient's 3D image data, a target mesh is constructed and data blocks are divided according to the aforementioned steps. The global maximum density value of the target volume is calculated, and a maximum density projection image is generated. During further processing, if no edge data regions not covered by complete data blocks are detected between the upper and lower boundary slices, the global maximum density value is directly used as the target global maximum density value for the target volume. In this case, there is no need to optimize the previously generated maximum density projection image. This operation avoids unnecessary complex calculations in the absence of edge data regions, simplifies the process, and effectively improves the efficiency of determining the target global maximum density value.
[0060] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing related hardware through computer-readable instructions. These computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. The aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, optical disk, or read-only memory (ROM), or random access memory (RAM).
[0061] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.
[0062] Further reference Figure 3 As a response to the above Figure 2 To implement the method shown, this application provides an embodiment of an image generation apparatus, which is similar to... Figure 2 Corresponding to the method embodiments shown, this device can be specifically applied to various electronic devices.
[0063] like Figure 3 As shown, the image generation device 400 of this embodiment includes: a construction module 401, a division module 402, a first determination module 403, a first acquisition module 404, a second determination module 405, and a generation module 406. Wherein: The construction module 401 is used to acquire three-dimensional image data and construct a target mesh based on the preset cutting direction for the three-dimensional image data; The partitioning module 402 is used to divide the three-dimensional image data into multiple data blocks based on the target grid and calculate the maximum density value of all voxels in each data block. The first determining module 403 is used to determine the maximum density value of all complete data blocks within the target volume of the three-dimensional image data based on the maximum density value of all voxels. The first acquisition module 404 is used to identify the boundary slices of the target volume in the projection direction and obtain the maximum density value of the boundary slices. The second determining module 405 is used to determine the global maximum density value of the target volume based on the maximum density value of the complete data block and the maximum density value of the boundary slice. The generation module 406 is used to generate a maximum density projection image of the target volume based on the global maximum density value, the maximum density value of the complete data block, and the maximum density value of the boundary slice.
[0064] This application embodiment constructs a target mesh to precisely divide 3D image data into multiple data blocks, and pre-calculates the maximum density value of voxels within each data block. During processing, it eliminates the need to compare each voxel traversed by the projection ray; instead, the operation is based on the maximum density value of the data block. First, the maximum density value of the complete data block within the target volume is determined, then the maximum density value of the boundary slices is identified, and finally, the global maximum density value is determined. This block-based processing method significantly reduces the computational load, avoiding the exponential growth in computational complexity caused by comparing voxels one by one in high-resolution 3D image data processing. Finally, a maximum density projection image is generated based on these pre-calculated and processed maximum density values, significantly improving image generation speed and efficiency.
[0065] In one embodiment, the first acquisition module 404 includes: The identification submodule is used to identify the upper and lower boundary slices of the target volume in the projection direction; The analysis submodule is used to analyze the matrices corresponding to the upper and lower boundary slices and obtain the position information of the matrices in the data block partitioning. The first acquisition submodule is used to extract the maximum density values of the upper and lower boundary slices based on the location information.
[0066] The embodiments of this application can avoid the tedious calculation of comparing all voxels of the entire three-dimensional volumetric image data one by one, greatly reducing the amount of computation, improving the generation speed of the maximum density projection image, and providing the required images for clinical diagnosis more quickly.
[0067] In one embodiment, the generation module 406 includes: The second acquisition submodule is used to acquire the voxel positions corresponding to the global maximum density value, the maximum density value of the complete data block, and the maximum density value of the boundary slice, respectively. The mapping submodule is used to generate a maximum density projection image of the target volume based on the voxel location.
[0068] This application embodiment accurately obtains voxel locations, avoiding the need for traditional methods to calculate all voxels one by one, significantly reducing the amount of computation, increasing the generation speed of maximum density projection images, and enabling the rapid provision of clear projection images of high-density structures such as brain blood vessels for clinical use.
[0069] In one embodiment, the image generation apparatus 400 further includes: The first information acquisition module is used to acquire storage hierarchy information of a multi-level storage system. The storage module is used to store the maximum density value of each data block in the target storage level of the hierarchical storage system based on storage hierarchy information.
[0070] This application embodiment stores the maximum density value and global maximum density value of each data block through a multi-level storage system, which reduces data reading time and greatly improves the computational efficiency when generating the maximum density projection image based on the global maximum density value, thus shortening the image generation time.
[0071] In one embodiment, the second determining module 405 includes: The third acquisition submodule is used to obtain the maximum density value of a complete data block from the multi-level storage system; The comparison submodule is used to compare the maximum density value of the complete data block with the maximum density value of the boundary slice. The determination submodule is used to determine the global maximum density value of the target volume based on the comparison results.
[0072] The embodiments of this application can quickly acquire data by utilizing a multi-level storage system, reducing data reading time, improving the efficiency of determining the global maximum density value, and thus accelerating the generation speed of the maximum density projection image.
[0073] In one embodiment, the image generation apparatus 400 further includes: The detection module is used to detect whether there are edge data regions between the upper and lower boundary slices that are not covered by the complete data block; The second acquisition module is used to acquire the maximum density value of the edge data region of the upper boundary slice and the maximum density value of the edge data region of the lower boundary slice, if they exist. The comparison module is used to compare the maximum density value of the edge data region of the upper boundary slice and the maximum density value of the edge data region of the lower boundary slice with the global maximum density value, respectively. The correction module is used to correct the global maximum density value based on the comparison results to obtain the target global maximum density value of the target volume. The optimization module optimizes the maximum density projection image based on the maximum density values of the edge data regions of the upper boundary slice, the maximum density values of the edge data regions of the lower boundary slice, and the target global maximum density value.
[0074] The embodiments of this application can optimize the maximum density projection image, making the generated image edges clearer and the information more complete, thus effectively improving the image quality.
[0075] In one embodiment, the image generation apparatus 400 further includes: The third determination module is used to determine the global maximum density value as the target global maximum density value of the target volume if there is no edge data region in the boundary slice that is not covered by the complete data block.
[0076] The embodiments of this application can avoid unnecessary complex calculations in the case of no edge data regions, simplify the process, and effectively improve the efficiency of determining the target global maximum density value.
[0077] To address the aforementioned technical problems, embodiments of this application also provide a computer device. Please refer to [link / reference needed]. Figure 4 , Figure 4 This is a basic structural block diagram of the computer device in this embodiment.
[0078] Computer device 6 includes a memory 61, a processor 62, and a network interface 63 that are interconnected via a system bus. It should be noted that only computer device 6 with memory 61, processor 62, and network interface 63 is shown in the figure; however, it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented alternatively. Those skilled in the art will understand that the computer device described herein is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes, but is not limited to, microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.
[0079] Computer devices can include desktop computers, laptops, handheld computers, and cloud servers. These devices allow for human-computer interaction with users through keyboards, mice, remote controls, touchpads, or voice-activated devices.
[0080] The memory 61 includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 61 may be an internal storage unit of the computer device 6, such as the hard disk or memory of the computer device 6. In other embodiments, the memory 61 may also be an external storage device of the computer device 6, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the computer device 6. Of course, the memory 61 may include both the internal storage unit and the external storage device of the computer device 6. In this embodiment, the memory 61 is typically used to store the operating system and various application software installed on the computer device 6, such as computer-readable instructions for image generation methods. In addition, the memory 61 may also be used to temporarily store various types of data that have been output or will be output.
[0081] In some embodiments, processor 62 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip. Processor 62 is typically used to control the overall operation of computer device 6. In this embodiment, processor 62 is used to execute computer-readable instructions stored in memory 61 or to process data, such as computer-readable instructions for executing an image generation method.
[0082] The network interface 63 may include a wireless network interface or a wired network interface, which is typically used to establish a communication connection between the computer device 6 and other electronic devices.
[0083] This application embodiment constructs a target mesh to precisely divide 3D image data into multiple data blocks, and pre-calculates the maximum density value of voxels within each data block. During processing, it eliminates the need to compare each voxel traversed by the projection ray; instead, the operation is based on the maximum density value of the data block. First, the maximum density value of the complete data block within the target volume is determined, then the maximum density value of the boundary slices is identified, and finally, the global maximum density value is determined. This block-based processing method significantly reduces the computational load, avoiding the exponential growth in computational complexity caused by comparing voxels one by one in high-resolution 3D image data processing. Finally, a maximum density projection image is generated based on these pre-calculated and processed maximum density values, significantly improving image generation speed and efficiency.
[0084] This application also provides another embodiment, namely, providing a computer-readable storage medium storing computer-readable instructions that can be executed by at least one processor to cause the at least one processor to perform the steps of the image generation method described above.
[0085] This application embodiment constructs a target mesh to precisely divide 3D image data into multiple data blocks, and pre-calculates the maximum density value of voxels within each data block. During processing, it eliminates the need to compare each voxel traversed by the projection ray; instead, the operation is based on the maximum density value of the data block. First, the maximum density value of the complete data block within the target volume is determined, then the maximum density value of the boundary slices is identified, and finally, the global maximum density value is determined. This block-based processing method significantly reduces the computational load, avoiding the exponential growth in computational complexity caused by comparing voxels one by one in high-resolution 3D image data processing. Finally, a maximum density projection image is generated based on these pre-calculated and processed maximum density values, significantly improving image generation speed and efficiency.
[0086] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods of the various embodiments of this application.
[0087] Obviously, the embodiments described above are only some embodiments of this application, not all embodiments. The accompanying drawings show preferred embodiments of this application, but do not limit the patent scope of this application. This application can be implemented in many different forms; rather, the purpose of providing these embodiments is to provide a more thorough and comprehensive understanding of the disclosure of this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing specific embodiments, or make equivalent substitutions for some of the technical features. Any equivalent structures made using the content of this application's specification and drawings, directly or indirectly applied to other related technical fields, are similarly within the scope of patent protection of this application.
[0088] The software tools or components not belonging to our company that appear in the embodiments of this application are merely examples and do not represent actual use.
Claims
1. An image generation method, characterized in that, Includes the following steps: Acquire 3D image data, and construct a target mesh based on the preset cutting direction for the 3D image data; Based on the target grid, the three-dimensional image data is divided into multiple data blocks, and the maximum density value of all voxels in each data block is calculated. Based on the maximum density value of all voxels, determine the maximum density value of all complete data blocks within the target volume of the three-dimensional image data; Identify the boundary slices of the target volume in the projection direction and obtain the maximum density value of the boundary slices; Based on the maximum density value of the complete data block and the maximum density value of the boundary slice, the global maximum density value of the target volume is determined; Based on the global maximum density value, the maximum density value of the complete data block, and the maximum density value of the boundary slice, a maximum density projection image of the target volume is generated.
2. The image generation method according to claim 1, characterized in that, The boundary slice includes an upper boundary slice and a lower boundary slice. The step of identifying the boundary slice of the target volume in the projection direction and obtaining the maximum density value of the boundary slice specifically includes: Identify the upper boundary slice and the lower boundary slice of the target volume in the projection direction; Analyze the matrices corresponding to the upper boundary slice and the lower boundary slice to obtain the position information of the matrices in the data block partitioning; Based on the location information, the maximum density values of the upper boundary slice and the lower boundary slice are extracted.
3. The image generation method according to claim 1, characterized in that, The step of generating the maximum density projection image of the target volume based on the global maximum density value, the maximum density value of the complete data block, and the maximum density value of the boundary slice specifically includes: Obtain the voxel positions corresponding to the global maximum density value, the maximum density value of the complete data block, and the maximum density value of the boundary slice, respectively. Based on the voxel positions, a maximum density projection image of the target volume is generated.
4. The image generation method according to claim 1, characterized in that, After the step of dividing the 3D image data into multiple data blocks based on the target grid and calculating the maximum density value of all voxels in each data block, the method further includes: Obtain storage hierarchy information for a multi-level storage system; Based on the storage hierarchy information, the maximum density value of each data block is stored in the target storage hierarchy of the hierarchical storage system.
5. The image generation method according to claim 4, characterized in that, The step of determining the global maximum density value of the target volume based on the maximum density value of the complete data block and the maximum density value of the boundary slice specifically includes: Obtain the maximum density value of the complete data block from the multi-level storage system; The maximum density value of the complete data block and the maximum density value of the boundary slice are compared. Based on the comparison results, the global maximum density value of the target volume is determined.
6. The image generation method according to claim 1, characterized in that, The boundary slice includes an upper boundary slice and a lower boundary slice. After the step of generating the maximum density projection image of the target volume based on the global maximum density value, the maximum density value of the complete data block, and the maximum density value of the boundary slice, the method further includes: Detect whether there are edge data regions between the upper boundary slice and the lower boundary slice that are not covered by complete data blocks; If they exist, then obtain the maximum density value of the edge data region of the upper boundary slice and the maximum density value of the edge data region of the lower boundary slice; The maximum density values of the edge data regions of the upper boundary slice and the lower boundary slice are compared with the global maximum density value, respectively. Based on the comparison results, the global maximum density value is corrected to obtain the target global maximum density value of the target volume; The maximum density projection image is optimized based on the maximum density value of the edge data region of the upper boundary slice, the maximum density value of the edge data region of the lower boundary slice, and the target global maximum density value.
7. The image generation method according to claim 6, characterized in that, After the step of detecting whether there is an edge data region between the upper boundary slice and the lower boundary slice that is not covered by a complete data block, the method further includes: If the boundary slice does not contain any edge data regions not covered by the complete data block, then the global maximum density value is determined as the target global maximum density value of the target volume.
8. An image generation apparatus, characterized in that, include: A construction module is used to acquire three-dimensional image data and construct a target mesh based on the three-dimensional image data according to a preset cutting direction; The partitioning module is used to divide the three-dimensional image data into multiple data blocks based on the target grid, and calculate the maximum density value of all voxels in each data block; The first determining module is used to determine the maximum density value of all complete data blocks within the target volume of the three-dimensional image data based on the maximum density value of all voxels. The first acquisition module is used to identify the boundary slices of the target volume in the projection direction and acquire the maximum density value of the data block to which the boundary slices belong; The second determining module is used to determine the global maximum density value of the target volume based on the maximum density value of the complete data block and the maximum density value of the boundary slice; The generation module is used to generate a maximum density projection image of the target volume based on the global maximum density value, the maximum density value of the complete data block, and the maximum density value of the boundary slice.
9. A computer device, characterized in that, The method includes a memory and a processor, wherein the memory stores computer-readable instructions, and the processor executes the computer-readable instructions to implement the steps of the image generation method as described in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-readable instructions that, when executed by a processor, implement the steps of the image generation method as described in any one of claims 1 to 7.