Image processing method, electronic equipment, storage medium and product

Abstract

The invention discloses an image processing method, electronic equipment, a storage medium and a product, and the method comprises the steps: carrying out the response according to an image display instruction, and enabling a three-dimensional image containing a head structure to serve as a background image to be displayed; the brain tissue segmentation image of the three-dimensional image is displayed at the corresponding position of the three-dimensional image in an overlapped mode, and display setting parameters during display of the brain tissue segmentation image and the three-dimensional image are different at least in display color and / or display transparency.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular to an image processing method, electronic device, storage medium and product. Background Technology

[0002] Magnetic resonance imaging (MRI) is an important medical imaging technique widely used for observing brain structures and analyzing lesions. With the development of artificial intelligence (AI) and deep learning, AI-based MRI image processing methods have achieved significant improvements in accuracy and are widely used in tasks such as automated segmentation and feature extraction.

[0003] In related technologies, deep learning models are typically used to segment the brain from MRI images and display the segmentation results. However, presenting the brain segmentation results relies on manual delineation, which is time-consuming and labor-intensive. Summary of the Invention

[0004] This application provides an image processing method, electronic device, storage medium, and product that solves the problem in related technologies that rely on manual outlining when presenting brain segmentation results, which is time-consuming and labor-intensive.

[0005] The technical solution of this application is implemented as follows:

[0006] An image processing method, the method comprising:

[0007] Responding to the image display command, the three-dimensional image containing the head structure is displayed as the background image;

[0008] The brain tissue segmentation image of the three-dimensional image is superimposed and displayed at the corresponding position of the three-dimensional image, wherein the display settings parameters of the brain tissue segmentation image and the three-dimensional image are different, at least in terms of display color and / or display transparency.

[0009] An image processing apparatus, comprising:

[0010] The display unit is used to respond to image display instructions and display a three-dimensional image containing the head structure as a background image.

[0011] The display unit is used to overlay the brain tissue segmentation image of the three-dimensional image at the corresponding position of the three-dimensional image, wherein the display settings parameters of the brain tissue segmentation image and the three-dimensional image are different, at least in terms of display color and / or display transparency.

[0012] An electronic device includes: a processor and a memory for storing a computer program capable of running on the processor.

[0013] When the processor runs the computer program, it executes the steps of any of the above methods.

[0014] A storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the above methods.

[0015] A computer product includes a computer program that, when executed by a processor, implements the steps of any of the above methods.

[0016] This application provides an image processing method that responds to image display instructions by displaying a three-dimensional image containing the head structure as a background image; and then overlays a segmented image of brain tissue from the three-dimensional image onto the corresponding position in the three-dimensional image. The display settings for the segmented brain tissue image and the displayed three-dimensional image differ, at least in display color and / or display transparency. This application accurately aligns the segmented brain tissue image to the corresponding position of the brain structure in the three-dimensional image to ensure the accuracy of the overlay display. By setting different display settings, the brain tissue is made more prominent in the original three-dimensional image. This presentation method helps to clearly understand the brain structure and its relationship with surrounding tissues, improving the visualization effect. Attached Figure Description

[0017] Figure 1 A flowchart illustrating an image processing method provided in this application embodiment. Figure 1 ;

[0018] Figure 2 A flowchart illustrating an image processing method provided in this application embodiment. Figure 2 ;

[0019] Figure 3 A flowchart illustrating an image processing method provided in this application embodiment. Figure 3 ;

[0020] Figure 4 A flowchart illustrating an image processing method provided in this application embodiment. Figure 4 ;

[0021] Figure 5 A schematic diagram of a corrected three-dimensional image provided in an embodiment of this application;

[0022] Figure 6 A schematic diagram of a binarized 3D image after preliminary threshold segmentation, provided for an embodiment of this application;

[0023] Figure 7 A schematic diagram of a three-dimensional image after morphological erosion operation, provided for an embodiment of this application;

[0024] Figure 8 A schematic diagram of a three-dimensional image after morphological dilation operation, provided for an embodiment of this application;

[0025] Figure 9 A schematic diagram of a three-dimensional image after three-dimensional smoothing, provided for an embodiment of this application;

[0026] Figure 10 A schematic diagram of a brain segment extracted after connected region analysis, provided as an embodiment of this application;

[0027] Figure 11 A schematic diagram showing the final segmented brain region displayed in superimposed colors, as provided in an embodiment of this application;

[0028] Figure 12 This is a schematic diagram of the structure of an image processing apparatus provided in an embodiment of this application;

[0029] Figure 13 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0030] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings.

[0031] It should be understood that the phrases "embodiments of this application" or "foreign embodiments" throughout the specification mean that a specific feature, structure, or characteristic related to an embodiment is included in at least one embodiment of this application. Therefore, "embodiments of this application" or "in the foreign embodiments" appearing throughout the specification do not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. In the various embodiments of this application, the sequence numbers of the above-described processes do not imply a sequential order of execution; the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application. The sequence numbers of the above-described embodiments are merely descriptive and do not represent the superiority or inferiority of the embodiments.

[0032] It should be noted that the terms "first, second, and third" used in the embodiments of this application are merely to distinguish similar objects and do not represent a specific ordering of objects. It is understood that "first, second, and third" can be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0033] In the following description, the terms three-dimensional image, brain tissue segmentation image, display setting parameters, etc., are key concepts for understanding the technical solution of this application, and are defined as follows:

[0034] 1) Three-dimensional images: These refer to three-dimensional medical imaging data containing head structures obtained through MRI. These images are usually stored in voxel format and can provide spatial distribution information of the internal structures of the human body.

[0035] 2) Brain tissue segmentation image: This refers to the portion of the image extracted from the original 3D image that represents only the brain region. This image is generated through a series of image processing steps (such as binarization, morphological operations, connected component analysis, etc.) to highlight brain tissue.

[0036] 3) Display settings parameters: These refer to parameters that control the image display effect during visualization, including but not limited to color and transparency. In this application, by setting different display settings parameters, a visual difference is created between the brain tissue segmentation image and the background 3D image, thereby improving the visualization effect.

[0037] In related technologies, traditional MRI brain image segmentation methods rely on manual annotation or complex deep learning models, which are not only time-consuming and labor-intensive but also require substantial computational resources, making it difficult to achieve fast and intuitive visualization results. To address these issues, this application provides a brain image segmentation and visualization method based on morphological processing and overlay display. This method achieves efficient and automatic segmentation of brain tissue through steps such as geometric correction, threshold segmentation, morphological refinement, connected component analysis, and image overlay display. By setting different display parameters, the segmentation results can be clearly and intuitively presented on top of the original image. The technical solution of this application has the advantages of simple operation, low computational resource consumption, and strong interactivity, and is suitable for multiple scenarios such as medical image analysis and scientific research applications.

[0038] The image processing methods provided in the embodiments of this application can be executed by a medical image processing system, which typically includes an image input unit, an image processing unit, and a display unit. Figure 1 This is an optional flowchart illustrating the image processing method provided in the embodiments of this application. The following will be combined with... Figure 1 The steps shown are explained as follows: Figure 1 As shown, the method includes the following steps S101 to S102:

[0039] Step S101: Respond to the image display instruction and display the three-dimensional image containing the head structure as the background image.

[0040] The image display command referred to here is a control signal issued by the user or system to trigger image display operations. Examples include a user clicking a "display image" button, calling a specific Application Programming Interface (API), or initiating a 3D image visualization process via voice recognition. Image display commands can originate from a graphical user interface (GUI), command-line terminal, remote control system, etc. The content of the image display command may include parameters such as the data path of the 3D image to be displayed, the 3D image type (e.g., fluid attenuated inversion recovery (FLAIR) sequence MRI), and the display mode (e.g., 2D slice, 3D reconstruction).

[0041] Three-dimensional images refer to three-dimensional medical imaging data containing head structures acquired through MRI. These images are stored in voxel format and provide spatial distribution information of the internal structures of the human body. This type of image data is typically in DICOM format and has multiple slices, each corresponding to a different Z-axis position. In this embodiment, the three-dimensional image can be either the original loaded MRI data or data that has undergone geometric correction, i.e., the distortion of the three-dimensional image caused by equipment errors or patient movement is eliminated through coordinate mapping and spatial transformation.

[0042] Background image refers to the three-dimensional image displayed as the underlying layer during visualization. In this application, the background image serves the basic function of displaying anatomical structures, while the brain tissue segmentation image is overlaid on the background image to create a contrast effect, facilitating the doctor's observation of brain regions.

[0043] In practice, when a user initiates an image display request, the system first loads the corresponding 3D MRI image data from the database or local storage, and performs geometric correction on the 3D image as needed. Subsequently, the system initializes the display module, renders the corrected 3D image into a complete 3D model, and sets the corrected 3D image as the background of the current display window. This 3D image rendering and display process can be implemented using a 3D graphics library, supporting multi-view observation and interactive zooming and rotation.

[0044] Step S102: The brain tissue segmentation image of the three-dimensional image is superimposed and displayed at the corresponding position of the three-dimensional image. The display settings parameters of the brain tissue segmentation image and the three-dimensional image are different, at least in terms of display color and / or display transparency.

[0045] Brain tissue segmentation images refer to the portion of an image extracted from an original 3D image that represents only the brain region.

[0046] The corresponding position refers to the location in 3D space where the brain tissue segmentation image corresponds to the location of the brain structure in the original 3D image. After segmentation, each non-zero voxel in the brain tissue segmentation image should correspond to the voxel at the same position in the original 3D image. This setting ensures that there will be no misalignment when overlaying the images.

[0047] Display settings parameters refer to a set of attributes used to control the display effect of an image during visualization, including but not limited to color, transparency, brightness, contrast, and texture mapping. In this application, to make brain tissue stand out more in the original 3D image, there is at least a difference in color and / or transparency between the brain tissue segmentation image and the background image. For example, the brain tissue segmentation image can be set to red with a certain degree of transparency (e.g., 50%), while the background image remains gray or white and is completely opaque, thus creating a sharp visual contrast.

[0048] Overlay display refers to the process of combining two or more images within the same display area according to certain rules. In this embodiment, a brain tissue segmentation image is overlaid onto a background 3D image in an overlay manner. This overlay operation makes the brain structure clearly visible against the original anatomical background. The image overlay process can be achieved using alpha blending techniques, for example, by weighting the pixels in the foreground image and the pixels in the background 3D image according to a transparency parameter, thereby generating the final display effect.

[0049] In practice, after displaying the background image, the system loads the pre-processed brain tissue segmentation image and renders it according to preset display settings (e.g., red color, 50% transparency). Then, the system overlays the brain tissue segmentation image onto the background image and adjusts its position to align it with the original brain structure. Users can further adjust parameters such as color, transparency, and viewing angle through the interactive interface for optimal viewing experience. Related technologies often lack intuitiveness and interactivity when presenting segmentation results, making it difficult for users to clearly understand the brain structure and its relationship with surrounding tissues.

[0050] The image processing method provided in this application, by setting different display parameters, makes brain tissue stand out more in the original three-dimensional image. This image processing method not only achieves automatic segmentation of brain tissue but also enhances visualization through differentiated settings of color and transparency, providing clinicians with an intuitive and efficient auxiliary tool.

[0051] In practice, steps S101 and S102 are closely related. First, the system must ensure that the background image is correctly loaded and displayed, which is the basis for subsequent brain tissue segmentation image overlay. Second, the brain tissue segmentation image is accurately aligned to the corresponding position of the brain structure in the original 3D image to ensure the accuracy and visual effect of the overlay display. The entire process coordinates multiple stages such as image loading, geometric correction, segmentation processing, and visualization rendering to ultimately achieve highlighted display of brain tissue in complex anatomical structures, thereby improving the efficiency and accuracy of medical image interpretation.

[0052] This application provides an image processing method that can overlay the segmentation results onto the original image without requiring complex computing resources.

[0053] In summary, this embodiment provides an image processing method with a clear structure and convenient operation. This image processing method is applicable to a variety of medical imaging scenarios and has significant application value in the field of neuroimaging.

[0054] In some embodiments, prior to step 102, such as Figure 2 As shown, steps 201-203 are included, wherein:

[0055] Step 201: Acquire a three-dimensional image, wherein the three-dimensional image includes a three-dimensional magnetic resonance imaging image after binarization.

[0056] Three-dimensional images refer to medical image data containing spatial location information. Each voxel has a grayscale value and corresponds to a point in a three-dimensional coordinate system. Three-dimensional images are typically acquired by MRI equipment and stored in DICOM format. In this embodiment, the three-dimensional image can be obtained by binarization after geometric correction, or it can be obtained by direct binarization without geometric correction. For example, in FLAIR sequence MRI images, there is a significant difference in intensity between brain tissue and surrounding tissue; therefore, the system can extract the brain region by setting an appropriate intensity threshold.

[0057] By acquiring 3D images through the system, the system can provide basic input for subsequent morphological processing and segmentation operations, thereby ensuring data consistency and accuracy during the processing.

[0058] Step 202: Perform morphological processing on the 3D image to obtain the processed 3D image.

[0059] Morphological processing is a structuring element-based operation method that can improve the shape and boundaries of target objects in an image while removing noise or small-sized interference. In this embodiment, morphological processing includes two steps: erosion and dilation. First, erosion is used to remove isolated small volumetric noise points, making the edges of the main target clearer. Then, dilation is used to restore the target area that has shrunk due to erosion, and it can also fill in any holes that may exist within the target area. Morphological processing helps improve image quality and makes subsequent connected component analysis more reliable.

[0060] By performing morphological processing on 3D images, the system can effectively remove unnecessary interference information and improve the accuracy and stability of the results obtained after segmenting 3D images.

[0061] Step 203: Perform connected component analysis on the processed 3D image to obtain a segmented brain tissue image.

[0062] Connected component analysis is an image processing technique used to identify interconnected sets of pixels in an image and classify them based on their geometric properties (such as volume and surface area). In this embodiment, connected component analysis is performed on the morphologically processed 3D image to identify all independent connected components and calculate the volume of each component. The system selects the connected component with the largest volume as the brain tissue segmentation image, while smaller connected components are excluded.

[0063] By using connected component analysis, the region that best matches the actual brain structure can be effectively selected from multiple candidate regions, thereby improving the accuracy and robustness of image segmentation results.

[0064] Brain tissue segmentation images are generated through a series of image processing steps, including intensity-threshold-based binarization segmentation, morphological refinement (such as erosion, dilation, and smoothing filtering), and connected component analysis. The final segmentation result is a 3D binary mask image containing only brain voxels, used to highlight brain structures.

[0065] In this embodiment, the system acquires three-dimensional images and performs morphological processing and connected component analysis on these images to achieve high-precision automatic segmentation of brain tissue. By performing the aforementioned processing on the three-dimensional images, the system removes noise and interference, thereby improving the accuracy of brain tissue segmentation images. This allows the system to provide high-quality basic data support for subsequent visualization and clinical applications.

[0066] In practice, the above steps have a clear sequential dependency. First, a high-quality 3D image must be obtained as the foundation for subsequent processing; second, morphological processing can further optimize the image quality, providing a more reliable input for connected component analysis; finally, based on the optimized image, connected component analysis completes the accurate segmentation of the target region. These steps together constitute a complete image preprocessing and segmentation method, ensuring the stability and accuracy of the segmentation results.

[0067] In summary, this embodiment provides a brain tissue segmentation method based on three-dimensional images. By acquiring three-dimensional images and performing morphological processing and connected component analysis, the method provided in this embodiment achieves efficient and accurate segmentation of brain structures.

[0068] In some embodiments, step 201: acquiring a three-dimensional image includes steps 2011-2013.

[0069] in:

[0070] Step 2011: Obtain a 3D image containing the head structure.

[0071] 3D images are raw, loaded 3D MRI images used for medical imaging analysis of brain structures. 3D images contain complete anatomical information of the head, including brain tissue, skull, blood vessels, and other relevant structures. The image data is typically in DICOM format, which facilitates reading and processing by medical image processing systems.

[0072] By acquiring high-quality 3D MRI images as input, a precise anatomical basis can be provided for subsequent segmentation and visualization operations. This method ensures that the segmented brain regions are highly accurate, thus providing reliable reference data for doctors or researchers.

[0073] Step 2012: Perform geometric correction on the 3D image to obtain the corrected 3D image.

[0074] Geometric correction refers to the spatial transformation of an original 3D image to conform to a standard coordinate system or to correct deformations caused by equipment errors. Geometric correction methods include affine transformation and non-rigid registration. After geometric correction, the voxels of the image are arranged more regularly, which is beneficial for subsequent image processing and analysis.

[0075] By performing geometric correction on medical images, image distortion caused by scanning equipment errors or patient displacement can be eliminated, thereby improving the accuracy of subsequent image segmentation and 3D visualization processes. After completing the above geometric correction, it can be ensured that the brain structures extracted from medical images have good spatial consistency.

[0076] Step 2013: Based on the set intensity threshold, binarize the corrected 3D image to obtain the 3D image.

[0077] Binarization is a process that converts a grayscale image into a black and white image. By setting an intensity threshold, pixels above the threshold are designated as foreground (usually white), and pixels below the threshold are designated as background (usually black). In the binarization process, the system uses the set intensity threshold to process the image, thereby initially locating brain regions and excluding most non-brain structures.

[0078] By setting an appropriate intensity threshold and performing binarization, image processing systems can coarsely separate brain regions in the early stages, reducing the computational load of subsequent processing and improving overall processing efficiency. This operation of setting an appropriate intensity threshold and performing binarization lays the foundation for subsequent morphological refinement and target extraction, thereby enabling a more efficient brain segmentation process.

[0079] In this embodiment, by acquiring a three-dimensional image containing the head structure and performing geometric correction and binarization, image distortion can be effectively removed and the brain region can be preliminarily separated. This improves the accuracy of subsequent segmentation and visualization.

[0080] In practice, a 3D image containing the head structure refers to a three-dimensional image obtained through magnetic resonance imaging (MRI) during medical image acquisition that encompasses the complete anatomical structure of the head. This 3D image includes not only the brain parenchyma but may also include structures such as the skull, cerebrospinal fluid, blood vessels, and soft tissues. 3D images containing the head structure are typically stored in DICOM format, possessing high resolution and rich grayscale information, making them an important data source for medical image processing and analysis. For example, in clinical practice, 3D images containing the head structure are frequently used in areas such as brain tumor detection, research on neurodegenerative diseases, and preoperative planning.

[0081] In summary, this embodiment achieves preprocessing and preliminary segmentation of the original image by sequentially acquiring a 3D image containing the head structure, performing geometric correction, and then binarizing it by setting an intensity threshold. These steps are interdependent and progressively advanced, first ensuring image quality, then eliminating spatial errors, and finally improving the efficiency of subsequent processing by simplifying the image structure. The entire process lays a solid foundation for subsequent fine segmentation and 3D visualization.

[0082] In some embodiments, step 202: performing morphological processing on the three-dimensional image to obtain a processed three-dimensional image, including steps 2021-2024, wherein:

[0083] Step 2021: Define the spherical three-dimensional structural element.

[0084] A spherical 3D structuring element is a fundamental geometric template for morphological operations. It is spherical in shape with a radius of three units. In three-dimensional space, it simulates neighborhood relationships to influence the boundaries of target regions during morphological operations such as erosion and dilation. Employing spherical 3D structuring elements allows for better adaptation to the curved surface characteristics of brain tissue, avoiding edge distortion or artifacts caused by using discontinuous shapes such as cubes.

[0085] By defining spherical three-dimensional structural elements, the geometric features of brain structures can be matched more naturally; this operation can improve the accuracy and fidelity of 3D images after morphological processing.

[0086] Step 2022: Based on the spherical three-dimensional structural element, perform an erosion operation on the three-dimensional image to obtain the eroded image.

[0087] Erosion is a morphological processing method that removes isolated small noise and shrinks the boundaries of the target region by sliding a spherical 3D structuring element across a 3D image and retaining only areas that are foreground in all positions. In this embodiment, the erosion operation utilizes a spherical 3D structuring element to remove discontinuous small noise patches in the 3D image, while maintaining the structural integrity of the main target.

[0088] By performing erosion operations, the system can remove small volumetric noise from 3D images, allowing subsequent image processing to focus more on the core target region and thus improve the robustness of image segmentation results.

[0089] Step 2023: Perform a dilation operation on the eroded image to obtain the dilated image.

[0090] Dilation is the inverse operation of erosion. It expands the target region and fills in small holes by setting all areas covered by a spherical 3D structuring element as the foreground. Dilation helps restore the main target region that was shrunk by erosion and connects potentially broken parts, resulting in a more complete overall structure in the image.

[0091] Dilation operations can restore the original size of the main target and repair internal holes, thereby enhancing the connectivity of the main target region and improving the quality of image segmentation.

[0092] Step 2024: Perform 3D smoothing on the dilated image to obtain the processed 3D image.

[0093] 3D smoothing refers to filtering a 3D image in three-dimensional space to reduce surface roughness and improve visual continuity. In this embodiment, Gaussian filtering is preferably used for 3D smoothing. The system performs a weighted average smoothing operation on each voxel and its neighborhood to achieve a smoother and more natural segmentation result.

[0094] 3D smoothing can further optimize the surface details of the segmentation results, making the segmentation results closer to the real brain structure, and improving the visualization effect.

[0095] In practical implementation, the spherical 3D structural element is indeed spherical with a radius of three units. Spherical 3D structural elements possess excellent isotropic properties, making them suitable for processing complex curved surface structures in 3D medical imaging. For example, when processing 3D brain MRI images, spherical 3D structural elements can better conform to curved structures such as brain sulci and gyri, thus avoiding local distortions caused by inappropriate structural element shape.

[0096] In practice, an erosion operation is performed on the binarized 3D image. Erosion removes small, isolated objects and shrinks the boundaries of the main targets. For example, when processing 3D images of brain segments with a lot of minute noise, erosion can effectively remove this noise while preserving the contour information of the main brain regions.

[0097] In practice, a dilation operation is performed on the eroded image. This dilation operation restores the size of the main target that has been reduced by erosion and fills in any small holes that may exist within the main target. For example, after erosion, some previously continuous brain regions may become broken; the dilation operation can reconnect these broken areas, thus making the three-dimensional image structure more complete.

[0098] In practical implementation, the dilated binarized 3D image undergoes 3D smoothing processing, such as using Gaussian filtering, to obtain a smoother brain surface. Here, a Gaussian filter is used to perform weighted averaging on voxels, which can significantly reduce surface roughness without changing the main structure of the brain, thus making the segmentation results more consistent with anatomical features.

[0099] In practice, the above steps are closely related and synergistic. First, defining suitable spherical 3D structural elements ensures good adaptability for subsequent morphological operations. Next, erosion removes noise and shrinks boundaries, followed by dilation to restore the integrity of the target region. Finally, 3D smoothing further optimizes the 3D image surface, resulting in a smoother and more natural final segmentation result. The entire process is interconnected, progressively improving segmentation quality and providing a reliable data foundation for subsequent medical analysis.

[0100] In summary, in this embodiment, by defining spherical three-dimensional structural elements and sequentially performing erosion, dilation, and three-dimensional smoothing processes, noise can be effectively removed, the target region restored, and the smoothed three-dimensional image obtained. This improves the accuracy and stability of brain segmentation, thereby enhancing the visualization of medical images.

[0101] In some embodiments, step 203: performing connected component analysis on the processed 3D image to obtain a brain tissue segmentation image, including steps 2031-2033, wherein:

[0102] Step 2031: Perform connected component analysis on the processed 3D image to identify all independent connected components.

[0103] Connected component analysis is an image processing technique used to identify all interconnected foreground regions (i.e., directly adjacent pixels or voxels) in a binary image. In this application, the processed 3D image has undergone morphological operations to remove noise and smooth boundaries. At this point, connected component analysis can identify all isolated, unconnected foreground regions, which are referred to as connected components. Each connected component represents a possible target structure, such as brain tissue or other non-target structures.

[0104] Through connected component analysis, the system can distinguish different structures in an image into multiple connected components, providing a foundation for subsequent selection of connected components that truly represent the brain. This avoids misidentifying other small structures or noise as brain tissue, thereby improving segmentation accuracy.

[0105] Step 2032: Obtain the geometric attribute data of each connected component.

[0106] Geometric attribute data refers to quantitative parameters describing the shape and size of connected components, such as volume, surface area, length, width, height, and center point coordinates. This application primarily focuses on the volume attribute. The volume of each connected component can be determined by calculating the number of voxels it contains. Volume is one of the important criteria for determining whether a connected component is brain tissue, because the brain, as one of the largest organs in the human body, typically has a significantly larger volume than other structures.

[0107] By acquiring the geometric attribute data of each connected component, especially its volume information, objective criteria can be provided for subsequent screening, thereby more accurately identifying the connected components most likely to represent the brain. This method of acquiring the geometric attribute data of each connected component and using volume information for screening avoids misjudgment caused by similar shapes but large volume differences, improving the reliability of the segmentation results.

[0108] Step 2033: Select the image of the region where the geometric attribute data is greater than the attribute threshold as the brain tissue segmentation image.

[0109] The attribute threshold is a preset value used to filter connected components that meet specific conditions. In this application, the attribute threshold is typically set to a volume value. Only when the volume of a connected component exceeds the attribute threshold will the system select that connected component as part of the brain tissue segmentation image. For example, if the attribute threshold is set to 1000 voxels, only connected components with a volume greater than 1000 voxels will be recognized by the system and considered to belong to brain tissue.

[0110] By setting attribute thresholds and filtering qualified connected components, it is possible to effectively remove small connected components that may be noisy or irrelevant structures, while retaining large connected components that are most likely to represent the brain. This ensures that the final segmented image contains only brain-related structures, thereby improving segmentation accuracy and usability.

[0111] Geometric attribute data includes volume. When the system evaluates the geometric attribute data, if the geometric attribute data of a connected component is greater than a set attribute threshold, then that connected component may be determined as the connected component with the largest volume.

[0112] In this application, geometric attribute data is primarily used to evaluate the physical properties of connected components to aid in the identification of brain structures. Volume is one of the most critical geometric attributes because it directly reflects the scale of the connected component. The attribute threshold is a screening criterion set according to the actual application scenario, used to filter out small-volume structures that do not meet the criteria. For example, in most cases, the brain is the largest continuous structure in an MRI image; therefore, selecting the largest connected component as the constituent part of the brain tissue segmentation image is reasonable and effective.

[0113] By using volume as the primary geometric attribute and combining it with attribute thresholding for filtering, brain structures can be extracted efficiently and accurately. This method not only simplifies the segmentation process but also improves the stability and robustness of the segmentation results, making it particularly suitable for the rapid and automated brain tissue segmentation needs of MRI images in clinical medicine.

[0114] In this embodiment, by performing connected component analysis on the processed 3D image and filtering based on geometric attribute data, the segmentation system can effectively identify and retain the largest connected components as brain tissue segmentation images. This method removes noise and irrelevant structures, thereby improving segmentation accuracy and ultimately achieving clearer and more reliable brain visualization.

[0115] In practical implementation, including volume in the geometric attribute data means that in this embodiment, the geometric attributes considered are primarily volume, rather than other attributes such as surface area, length, width, and height. Volume is a core indicator for measuring the spatial occupancy of connected components. In brain MRI images, brain tissue typically has a much larger volume than other structures. Therefore, using volume as the main content of geometric attribute data helps to more accurately distinguish brain tissue from other structures.

[0116] Furthermore, connected components with geometric attribute data greater than the attribute threshold can be the largest connected components in terms of volume. This indicates that in this embodiment, the set attribute threshold is not absolutely fixed but can be adjusted according to actual conditions. When the attribute threshold is set low, multiple connected components may meet the conditions, and among these components, the largest connected component usually most closely resembles the actual brain structure. Therefore, in practical applications, the largest connected component can be selected as a component of the final brain tissue segmentation image, thereby further improving the accuracy of segmentation.

[0117] In practice, the above steps are closely logically linked. First, the image is decomposed into multiple connected components through connected component analysis; then, the geometric attribute data of each connected component, especially its volume, is obtained to provide a quantitative basis for subsequent selection; finally, by setting attribute thresholds, the largest connected components most likely representing the brain are selected. This progressive processing method makes the entire segmentation process more systematic, reliable, and more suitable for the actual processing needs of complex medical images.

[0118] In some embodiments, step 102: the brain tissue segmentation image is overlaid and displayed at the corresponding position in the three-dimensional image, such as... Figure 3 As shown, steps 301-302 are included, wherein:

[0119] Step 301: In response to the image rotation command, the brain tissue segmentation image and the 3D image are rotated synchronously.

[0120] Image rotation commands are control instructions issued by the user through an interactive interface, instructing the system to synchronously rotate the currently displayed 3D image and its segmentation results by a specified angle and direction. Image rotation commands can be triggered in various ways, such as mouse dragging, touch gestures, keyboard shortcuts, or virtual reality controllers. When the user executes the image rotation operation, the system analyzes the user's input in real time and converts it into specific rotation angle and axis information, thereby driving the 3D image and its segmentation results to rotate synchronously. This interactive mechanism allows users to observe the relationship between the segmented brain tissue image and the surrounding tissue in the 3D image from multiple perspectives, improving the flexibility and practicality of visualization.

[0121] Image rotation commands refer to the process by which users determine the rotation angle and direction based on their actions on the interactive interface (such as swiping, clicking, dragging, etc.), thereby rotating the image from different angles. For example, a user can hold down the left mouse button and drag a 3D image on the screen; the system calculates the corresponding rotation angle and axis based on the dragging direction and distance, thus achieving multi-angle rotation of the image. Furthermore, in a virtual reality environment, users can also control the rotation direction and angle of the image using a controller or head movement.

[0122] In this embodiment, the brain tissue segmentation image and the original 3D image are synchronously rotated in response to an image rotation command. By responding to the image rotation command and synchronously rotating the brain tissue segmentation image and the original 3D image, it can be ensured that the brain tissue segmentation image is always accurately superimposed on the 3D image. This method can thus provide a stable visual reference, which in turn can help doctors or researchers to more intuitively understand the spatial distribution and morphological changes of brain structures.

[0123] Step 302: The rotated brain tissue segmentation image is superimposed and displayed at the corresponding position in the rotated 3D image.

[0124] After completing the synchronous rotation, the system remaps the brain tissue segmentation image onto the rotated 3D background image based on the rotated coordinates, and displays it overlaid with preset colors and transparency. Since rotation affects the spatial relationships of the images, it is crucial to ensure pixel alignment between the brain tissue segmentation image and the background image to avoid offset or misalignment caused by rotation. The entire process of remapping the brain tissue segmentation image onto the rotated 3D background image typically involves image resampling and perspective transformation algorithms to ensure that the rotated brain tissue segmentation image maintains a consistent geometric correspondence with the background 3D image.

[0125] In practice, there is a close relationship between image rotation commands and the resulting image overlay. That is, the system will only execute the corresponding rotation operation after receiving a clear image rotation command, and will proceed with subsequent image overlay processing only after the rotation operation is completed. If no valid image rotation command is received, the system will maintain the default viewpoint and image layout.

[0126] In this embodiment of the application, by superimposing the rotated brain tissue segmentation image at the corresponding position of the rotated three-dimensional image, it can be ensured that the brain tissue segmentation image maintains spatial consistency with the original three-dimensional image from any viewpoint, thereby improving the visualization quality of the image.

[0127] Image rotation commands, as a crucial component of interactive visualization, are essential for enhancing user experience. By translating user-inputted rotation intentions into precise rotation parameters and driving the system to synchronously update 3D images and segmented brain tissue images, the system can provide flexible multi-angle viewing capabilities while maintaining image clarity. This interactive approach not only enhances the usability of the visualization system but also provides a more intuitive operational experience for medical image analysis.

[0128] In actual implementation, the various steps of the entire embodiment have a clear logical sequence and collaborative relationship. First, the system receives the image rotation command issued by the user. Then, the system synchronously rotates the 3D image and the brain tissue segmentation image according to the image rotation command. Finally, the system overlays the rotated brain tissue segmentation image onto the rotated 3D image. The above process ensures that the user maintains control over the spatial relationships of the images throughout the operation, while improving the interactivity and visualization of medical image analysis.

[0129] In some embodiments, after step 102, the method further includes the following steps: in response to an image scaling instruction, synchronously scaling the brain tissue segmentation image and the three-dimensional image, and superimposing the scaled brain tissue segmentation image at the corresponding position in the scaled three-dimensional image.

[0130] Image scaling commands are user-input instructions via an interactive interface to adjust the size of an image. These commands can be triggered by mouse wheel movements, touchscreen pinch gestures, keyboard shortcuts, or other methods. They typically include scaling information to control the degree to which the image is magnified or reduced. For example, scrolling the mouse wheel upwards is recognized as a zoom-in command, while scrolling downwards is recognized as a zoom-out command.

[0131] Synchronous scaling refers to the simultaneous magnification or reduction of a segmented brain tissue image and the original 3D image at the same scale when the user performs an image scaling operation. Synchronous scaling ensures that the spatial correspondence between the segmented brain tissue image and the original 3D image is not disrupted by scaling. Synchronous scaling is crucial for maintaining the accuracy of visualizations, especially in the field of medical imaging, where doctors and researchers often need to observe the relationship between brain structures and surrounding tissues at different scales.

[0132] In practical applications, when a user issues an image zoom command, the system first parses the zoom ratio corresponding to the command. Then, the system applies the same zoom algorithm (such as bilinear interpolation, nearest neighbor interpolation, etc.) to both the brain tissue segmentation image and the 3D image. Finally, the system re-overlays the zoomed brain tissue segmentation image according to its relative position in the original 3D image. This processing method ensures that regardless of how the user zooms the image, the brain region remains in the correct position, facilitating observation and analysis.

[0133] Furthermore, the synchronized scaling feature supports various interaction methods, such as continuous scaling and smooth transitions, to enhance the user experience. For example, in the web application, by listening to mouse wheel events or touchpad gestures, the system can update the image scaling status in real time and dynamically update the displayed image content through the graphics rendering engine, thereby achieving smooth visual feedback.

[0134] In this embodiment, by responding to image scaling commands and synchronously scaling the brain tissue segmentation image and the 3D image, the relationship between brain structures and background tissues can be clearly displayed at different scales. This improves the flexibility and accuracy of image observation.

[0135] In practice, determining the scaling size refers to calculating the proportional parameters by which the image should be enlarged or reduced based on the user's input image scaling command. For example, when a user clicks the "zoom in 50%" button through the interactive interface, the system will multiply the current image display size by 1.5 to obtain the new display size. The system executes the process of determining the scaling size, which includes parsing the user input signal, calculating the scaling factor, and preparing for subsequent image redrawing.

[0136] In practice, enabling zoomable image viewing refers to presenting the scaled image at a new size on the user interface after the image scaling operation, allowing users to view image details at different scales. For example, a user might need to zoom in on a specific brain region to check if its boundaries are clear, or zoom out on the entire image to get an overview of the overall layout. The process of enabling zoomable image viewing is typically combined with graphics rendering techniques to ensure image clarity and stability at different scaling levels.

[0137] In summary, the steps described above are closely interconnected, collectively enabling interactive scaling and synchronous overlay display of images. Here, the user first inputs an image scaling command through the interactive interface. The system then determines the scaling size based on this command and synchronously scales the brain tissue segmentation image and the 3D image accordingly. Finally, the scaled brain tissue segmentation image is overlaid and displayed at the corresponding position in the scaled 3D image. In practice, these steps are typically integrated into an image processing and visualization system, achieving efficient and stable image interaction through front-end and back-end collaboration.

[0138] In this embodiment, the steps of the magnetic resonance imaging (MRI) brain image segmentation and visualization method are integrally mapped onto the hardware architecture of an electronic device, thereby automating the processing of the MRI brain image segmentation and visualization method. This integral mapping ensures the efficient execution of the MRI brain image segmentation and visualization method in practical applications. This mapping method improves the speed and accuracy of medical image processing, and consequently provides an intuitive and reliable auxiliary tool for clinical medical applications.

[0139] In summary, this embodiment achieves automated operation of the magnetic resonance imaging (MRI) brain image segmentation and visualization method by rationally mapping the steps and hardware architecture. This embodiment improves image processing efficiency and enhances the reliability of results through the aforementioned mapping method. The method and system provided in this embodiment offer strong support for subsequent clinical medical applications.

[0140] In this embodiment, the method involved in this application is stored in a computer-readable storage medium, enabling the method to be executed by a processor. This operation can be applied to the field of medical image processing, thereby improving the efficiency and accuracy of brain MRI image segmentation and visualization.

[0141] In a feasible scenario, such as Figure 4 As shown, a method flowchart for MRI brain image segmentation and visualization is provided.

[0142] Step 401: Load the three-dimensional magnetic resonance imaging data containing the head structure and perform geometric correction.

[0143] First: Obtain the raw 3D magnetic resonance imaging data;

[0144] Raw 3D MRI image data refers to unprocessed medical image data acquired using MRI equipment. Medical personnel store the raw 3D MRI image data in DICOM format, which contains multiple slice images and spatial location information of the raw 3D MRI image data. Medical technicians use the raw 3D MRI image data for subsequent segmentation and visualization operations. Common image types include T1-weighted, T2-weighted, and FLAIR sequences, each providing different tissue contrast.

[0145] Secondly, geometric correction is performed on the 3D magnetic resonance imaging data to correct image distortion and convert it to standard coordinate space.

[0146] Geometric correction refers to the spatial transformation of 3D magnetic resonance imaging (MRI) data to eliminate image distortions caused by equipment errors or human posture. Geometric correction ensures that 3D MRI data conforms to a unified standard coordinate system. The process involves mapping each voxel in the 3D MRI image data to a standardized space, facilitating subsequent processing and comparison. For example, using intrinsic parameter matrices for coordinate transformation, 3D MRI data acquired by different scanning devices can be analyzed within the same reference frame.

[0147] In practical applications, geometric correction can be implemented using software tools. For example, technicians can use open-source medical image processing libraries to perform automated geometric correction on 3D magnetic resonance imaging data to improve the consistency and comparability of such image data.

[0148] Step 402: Based on the intensity values ​​of multiple voxels in the three-dimensional magnetic resonance imaging data, set an intensity threshold and binarize the three-dimensional magnetic resonance imaging data to generate a preliminary binary mask of the brain candidate region.

[0149] An intensity threshold is a boundary point set based on the grayscale values ​​of voxels in 3D magnetic resonance imaging (MRI) data. It is used to distinguish between foreground (brain tissue) and background (non-brain tissue). The selection of the intensity threshold is based on the statistical characteristics of the 3D MRI data, such as histogram distribution, mean, and standard deviation. By setting a reasonable intensity threshold, brain regions can be initially separated, laying the foundation for subsequent processing.

[0150] Binarization methods based on intensity threshold settings are simple and efficient, and applicable to various MRI sequences, especially FLAIR sequences. FLAIR sequences have good cerebrospinal fluid suppression effects, which can enhance the contrast of brain tissue and improve segmentation accuracy.

[0151] Step 403: Define a three-dimensional structural element.

[0152] Three-dimensional structuring elements are templates used in morphological processing to guide the direction and extent of operations such as erosion and dilation. This embodiment uses spherical structuring elements with a radius of three units, representing a distance of three units extending in each of six directions around the central voxel in three-dimensional space. The shape and size of the structuring element affect the fineness of the processing result; spherical structuring elements help maintain smooth edges.

[0153] Step 404: Apply an erosion operation and a dilation operation sequentially to the binary mask of the preliminary brain candidate region.

[0154] Erosion is a fundamental morphological operation used to remove small, isolated noise points and shrink the boundaries of the target region. It works by sliding 3D structuring elements across the target region, retaining only voxels completely covered by these elements. Erosion effectively reduces interference from mis-segmentation and improves segmentation accuracy.

[0155] Dilation, in contrast to erosion, is used to expand the boundaries of the target area, restore the main target size that has been reduced by erosion, and fill any small holes that may exist inside. Dilation uses 3D structuring elements to slide within the background area, thereby expanding the target area outwards to compensate for the losses caused by erosion.

[0156] The order of these two operations cannot be reversed. The erosion operation is performed first, followed by the dilation operation. This order ensures that the main structure is preserved while removing noise, thus avoiding misidentification due to over-dilation.

[0157] In practice, erosion and dilation operations are closely related. Erosion is used to remove noise from the image, while dilation repairs structural defects caused by erosion. The combination of erosion and dilation forms the basis of morphological filtering, a crucial step in achieving high-quality image segmentation.

[0158] Step 405: Apply a three-dimensional smoothing filter to the result of the dilation operation to produce a refined binary mask.

[0159] 3D smoothing filtering is an image processing technique used to reduce surface irregularities in segmented regions. This embodiment employs Gaussian smoothing filtering. Gaussian smoothing filtering achieves edge smoothing by calculating the weighted average between each voxel and its neighboring voxels. The weights of the Gaussian filter follow a Gaussian distribution; voxels closer to the center have a greater impact on the smoothing result, while voxels farther from the center have a smaller impact. Therefore, Gaussian smoothing filtering can effectively reduce high-frequency noise while preserving key structural features.

[0160] The application of 3D smoothing filtering makes the final segmentation result smoother and more natural, improving visual continuity.

[0161] Step 406: Perform a connected component analysis on the refined binary mask to identify at least one connected component, and select one of the connected components as the final brain tissue segmentation result based on the default geometric properties of the at least one connected component.

[0162] Connected component analysis is used to identify and label interconnected regions in an image. In this embodiment, the system processes the refined binary mask using connected component analysis, enabling it to identify all independent connected components and further calculate the volume and other geometric properties of each component.

[0163] Since the brain is typically the largest connected component, this embodiment selects the largest connected component as the final brain segmentation result. This selection method is simple and effective, avoids the need for manual intervention, and improves the efficiency and consistency of automated processing.

[0164] Step 407: Display the original loaded 3D MRI image data or a geometrically corrected 3D MRI image data as a background image, and overlay the final brain tissue segmentation result onto the background image with a default color and a preset transparency that are different from the background image.

[0165] Overlay display involves superimposing segmented brain structures onto the original image with specific colors and transparency to enhance visualization. In this embodiment, the background image is the original or corrected MRI image, while the foreground image is the segmented brain region, typically marked with colors such as red, green, or blue. The system is set with appropriate transparency to allow observation of the brain structures and their relative positions within the original image.

[0166] Visualization is not only intuitive and clear, but it also allows doctors or researchers to observe and analyze brain structure from different angles.

[0167] In practical applications, the steps can be achieved through 3D volumetric rendering technology, which allows users to interactively rotate, scale, and observe images, providing users with richer perspectives and a more flexible operating experience.

[0168] In actual implementation, the various steps of the entire embodiment have a clear sequential dependency. First, the quality and consistency of the input data are ensured by loading and correcting the three-dimensional magnetic resonance imaging data. Next, the segmentation results are gradually optimized through intensity threshold setting and morphological operations, removing noise and preserving the main structures. Subsequently, the final segmentation result is determined through connected component analysis. Finally, the processing results are displayed through visualization overlay. The system executes the above steps sequentially, making them interconnected and forming a complete MRI brain image segmentation and visualization workflow.

[0169] In summary, this embodiment of the application achieves simple and efficient MRI brain image segmentation and visualization by combining threshold segmentation, morphological processing, connected component analysis, and visualization overlay techniques. This method, which combines threshold segmentation, morphological processing, connected component analysis, and visualization overlay techniques to achieve simple and efficient MRI brain image segmentation and visualization, can remove noise and optimize segmentation results, thereby improving the accuracy of brain structure extraction and providing intuitive and reliable image support.

[0170] This application provides a method and system for MRI brain image segmentation and visualization. The method first loads three-dimensional MRI image data of the head and performs image geometric correction. Next, the image is initially binarized by setting an intensity threshold to roughly locate brain regions. Subsequently, a series of morphological processing steps, including erosion, dilation, and three-dimensional smoothing, are employed to refine the segmentation results and remove noise. Further, connected component analysis is used to identify and retain the largest connected regions as the final brain tissue segmentation result. Finally, the segmented brain regions are overlaid on the original head image with specific colors and transparency to achieve a clear and intuitive visualization effect. This application aims to simply and effectively extract brain structures automatically from MRI images and provide enhanced visualization tools.

[0171] The main objective of this application is to provide a simple method for segmenting and visualizing MRI brain images. This method combines thresholding, morphological manipulation, connected component analysis, and overlay display techniques. By using these techniques, the brain can be separated from head MRI images and visualized in a simple and rapid manner.

[0172] In a feasible scenario, to achieve the above objectives, this application provides a method for segmenting and visualizing magnetic resonance imaging (MRI) brain images. The steps of this method include at least:

[0173] Image acquisition and preprocessing steps: The system loads 3D magnetic resonance imaging data containing the head structure; the system performs geometric correction on the loaded image data; and the system converts the image data to a standard coordinate space or corrects image distortion. For example, Figure 5 This is a schematic diagram of the corrected 3D image.

[0174] Preliminary segmentation step: The image processing system, based on the intensity values ​​of voxels in the image, sets a predetermined threshold and performs binarization processing on the 3D image to generate preliminary candidate brain regions. For example, Figure 6 This is a schematic diagram of the binarized 3D image after preliminary threshold segmentation.

[0175] Morphological refining steps:

[0176] A system or user defines a spherical three-dimensional structural element with a radius of three units.

[0177] In the image processing workflow, an erosion operation is applied to the initial binarized 3D image to remove isolated small noise points and further refine the edge contours of the target object. For example, Figure 7 This is a schematic diagram of a three-dimensional image after morphological erosion.

[0178] A dilation operation is applied to the eroded binarized 3D image to restore the size of the main target area and fill the internal small holes. For example, Figure 8 This is a schematic diagram of a 3D image after morphological dilation.

[0179] The operator applies a 3D smoothing filter to the dilated binarized 3D image to achieve smoothing of the surface of the segmented region. For example, Figure 9 This is a schematic diagram of a 3D image after 3D smoothing.

[0180] Target extraction steps:

[0181] The system performs connected component analysis on the morphologically refined binary 3D image and identifies all independent connected components.

[0182] Calculate the volume or other geometric properties of each connected component.

[0183] Image segmentation systems select the connected component with the largest volume as the final brain tissue segmentation result. For example, Figure 10 This is a schematic diagram of the brain extracted after connected component analysis.

[0184] Overlay visualization steps:

[0185] The original 3D head image is displayed as the background.

[0186] The image processing system uses the final segmented brain tissue as the foreground image and overlays it onto the background image using a preset color and transparency different from the background image. For example, Figure 11 The image shows a schematic diagram of the final segmented brain, displayed in superimposed colors.

[0187] For example, magnetic resonance imaging data can be FLAIR sequence images.

[0188] For example, a three-dimensional structuring element is set as a spherical structuring element, and its radius is set to three units.

[0189] For example, the 3D smoothing filter is set to Gaussian smoothing filter.

[0190] For example, the overlay visualization step utilizes 3D volumetric rendering technology, which allows users to observe the overlaid image from different perspectives.

[0191] Through the above technical solution, this application can easily extract brain structures from head MRI images and clearly display the segmented brain structures in the original head images by using color and transparency to distinguish them, thus providing a powerful tool for medical image analysis.

[0192] This application also provides an image processing apparatus, such as... Figure 12 As shown, the device includes:

[0193] The display unit 501 is used to respond to the image display instruction and display a three-dimensional image containing the head structure as a background image.

[0194] Display unit 501 is used to overlay brain tissue segmentation images of a three-dimensional image onto the corresponding positions of the three-dimensional image, wherein the display settings parameters for displaying the brain tissue segmentation image and the three-dimensional image are different, at least in terms of display color and / or display transparency.

[0195] In some embodiments, the device further includes: an image input unit 502 for acquiring a three-dimensional image, wherein the three-dimensional image includes a three-dimensional magnetic resonance imaging image after binarization.

[0196] The image processing unit 503 is used to perform morphological processing on the three-dimensional image to obtain the processed three-dimensional image.

[0197] The image processing unit 503 is used to perform connected component analysis on the processed three-dimensional image to obtain a brain tissue segmentation image.

[0198] In some embodiments, the image input unit 502 is used to acquire a three-dimensional magnetic resonance imaging image containing the head structure;

[0199] The image processing unit 503 is used to perform geometric correction on the three-dimensional magnetic resonance imaging image to obtain the corrected three-dimensional image.

[0200] The image processing unit 503 is used to perform binarization processing on the corrected three-dimensional image based on a set intensity threshold to obtain a three-dimensional image.

[0201] In some embodiments, the image processing unit 503 is used to define spherical three-dimensional structural elements;

[0202] Image processing unit 503 is used to perform erosion operation on a three-dimensional image based on a spherical three-dimensional structural element to obtain an eroded image;

[0203] Image processing unit 503 is used to perform a dilation operation on the eroded image to obtain a dilated image;

[0204] The image processing unit 503 is used to perform three-dimensional smoothing processing on the dilated image to obtain the processed three-dimensional image.

[0205] In some embodiments, the image processing unit 503 is used to perform connected component analysis on the processed 3D image to identify all independent connected components.

[0206] Image processing unit 503 is used to acquire geometric attribute data of each connected component;

[0207] Image processing unit 503 is used to select images of connected components whose geometric attribute data are greater than the attribute threshold as brain tissue segmentation images.

[0208] In some embodiments, the display unit 501 is configured to synchronously rotate the brain tissue segmentation image and the three-dimensional image in response to an image rotation command, and to overlay the rotated brain tissue segmentation image onto the corresponding position of the rotated three-dimensional image.

[0209] In some embodiments, the display unit 501 is configured to synchronously scale the brain tissue segmentation image and the three-dimensional image in response to an image scaling instruction, and to overlay the scaled brain tissue segmentation image onto the corresponding position of the scaled three-dimensional image.

[0210] Embodiments of this application provide an electronic device that can be applied to... Figure 1 In a corresponding embodiment of a video processing method, referring to Figure 13 As shown, the electronic device 600 includes: a display 601, a processor 602, a memory 603, and a communication bus 604, wherein:

[0211] The communication bus 604 is used to realize the communication connection between the display 601, the processor 602, and the memory 603.

[0212] Memory 603 is used to store computer programs that can run on processor 602;

[0213] When running a computer program, the processor 602 performs the following steps: controlling the display 601 to respond to an image display instruction and displaying a three-dimensional image containing the head structure as a background image; and overlaying a segmented brain tissue image of the three-dimensional image at a corresponding position in the three-dimensional image, wherein the display settings parameters for the segmented brain tissue image and the three-dimensional image are different, at least in terms of display color and / or display transparency.

[0214] In some embodiments, the processor 602 is configured to acquire a three-dimensional image, wherein the three-dimensional image includes a three-dimensional magnetic resonance imaging image after binarization; perform morphological processing on the three-dimensional image to obtain a processed three-dimensional image; and perform connected component analysis on the processed three-dimensional image to obtain a brain tissue segmentation image.

[0215] In some embodiments, the processor 602 is configured to acquire a three-dimensional magnetic resonance imaging image containing the head structure; perform geometric correction on the three-dimensional magnetic resonance imaging image to obtain a corrected three-dimensional image; and perform binarization processing on the corrected three-dimensional image based on a set intensity threshold to obtain a three-dimensional image.

[0216] In some embodiments, the processor 602 is configured to define spherical three-dimensional structural elements; perform an erosion operation on a three-dimensional image based on the spherical three-dimensional structural elements to obtain an eroded image; perform a dilation operation on the eroded image to obtain a dilated image; and perform three-dimensional smoothing processing on the dilated image to obtain a processed three-dimensional image.

[0217] In some embodiments, the processor 602 is configured to perform connected component analysis on the processed 3D image to identify all independent connected components; acquire geometric attribute data of each connected component; and select the image of the region where the connected component with geometric attribute data greater than the attribute threshold is located as the brain tissue segmentation image.

[0218] In some embodiments, the processor 602 is configured to control the display 601 to synchronously rotate the brain tissue segmentation image and the three-dimensional image in response to an image rotation command, and to overlay the rotated brain tissue segmentation image onto the corresponding position of the rotated three-dimensional image.

[0219] In some embodiments, the processor 602 is configured to control the display 601 to synchronously scale the brain tissue segmentation image and the three-dimensional image in response to an image scaling instruction, and to overlay the scaled brain tissue segmentation image onto the corresponding position of the scaled three-dimensional image.

[0220] A processor can be an integrated circuit chip with signal processing capabilities, such as a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, a general-purpose processor can be a microprocessor or any conventional processor.

[0221] It should be noted that the descriptions of the same steps and contents as in other embodiments in this embodiment can be found in the descriptions in other embodiments, and will not be repeated here.

[0222] Embodiments of this application provide a computer storage medium storing one or more programs, which can be executed by one or more processors to achieve, for example... Figure 1 The steps are shown.

[0223] It should be noted that the descriptions of the same steps and contents as in other embodiments in this embodiment can be found in the descriptions in other embodiments, and will not be repeated here.

[0224] It should be noted that the aforementioned computer storage media / memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), magnetic random access memory (FRAM), flash memory, magnetic surface memory, optical disc, or compact disc read-only memory (CD-ROM), etc.; it can also be various terminals that include one or any combination of the above-mentioned memory, such as mobile phones, computers, tablet devices, personal digital assistants, etc.

[0225] Embodiments of this application provide a computer product, including a computer program, which can be executed by a processor 602 of an electronic device 600 to perform tasks such as... Figure 1 The steps are shown.

[0226] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0227] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0228] Furthermore, in the various embodiments of this application, all functional units can be integrated into one processing module, or each unit can be a separate unit, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in a combination of hardware and software functional units. Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0229] The methods disclosed in the several method embodiments provided in this application can be arbitrarily combined without conflict to obtain new method embodiments.

[0230] The features disclosed in the several product embodiments provided in this application can be arbitrarily combined without conflict to obtain new product embodiments.

[0231] The features disclosed in the several method or device embodiments provided in this application can be arbitrarily combined without conflict to obtain new method or device embodiments.

[0232] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. An image processing method, characterized by, The method comprises: in response to the image display instruction, displaying a three-dimensional image containing head structure as a background image; superimposedly displaying a brain tissue segmentation image of the three-dimensional image at a corresponding position of the three-dimensional image, wherein the display setting parameters of the brain tissue segmentation image and the three-dimensional image when displayed are different at least in display color and / or display transparency.

2. The method of claim 1, wherein, Before the superimposedly displaying the brain tissue segmentation image of the three-dimensional image at the corresponding position of the three-dimensional image, the method comprises: obtaining the three-dimensional image, wherein the three-dimensional image comprises a three-dimensional magnetic resonance imaging image after binarization processing; performing morphological processing on the three-dimensional image to obtain a processed three-dimensional image; performing connected region analysis on the processed three-dimensional image to obtain the brain tissue segmentation image.

3. The method of claim 2, wherein, The obtaining the three-dimensional image comprises: obtaining a three-dimensional magnetic resonance imaging image containing head structure; performing geometric correction on the three-dimensional magnetic resonance imaging image to obtain a corrected three-dimensional image; based on a set intensity threshold, performing binarization processing on the corrected three-dimensional image to obtain the three-dimensional image.

4. The method of claim 2, wherein, The performing morphological processing on the three-dimensional image to obtain a processed three-dimensional image comprises: defining a spherical three-dimensional structure element; based on the spherical three-dimensional structure element, performing an erosion operation on the three-dimensional image to obtain an eroded image; performing a dilation operation on the eroded image to obtain a dilated image; performing three-dimensional smoothing processing on the dilated image to obtain the processed three-dimensional image.

5. The method of claim 2, wherein, The performing connected region analysis on the processed three-dimensional image to obtain the brain tissue segmentation image comprises: performing connected region analysis on the processed three-dimensional image to identify all independent connected components; obtaining geometric attribute data of each connected component; selecting an image of a region where a connected component with geometric attribute data greater than an attribute threshold is located as the brain tissue segmentation image.

6. The method according to any one of claims 1 to 5, characterized in that, After the superimposedly displaying the brain tissue segmentation image of the three-dimensional image at the corresponding position of the three-dimensional image, the method comprises: in response to an image rotation instruction, synchronously rotating the brain tissue segmentation image and the three-dimensional image, and superimposedly displaying the rotated brain tissue segmentation image at a corresponding position of the rotated three-dimensional image.

7. The method according to any one of claims 1 to 5, characterized in that, After the superimposedly displaying the brain tissue segmentation image of the three-dimensional image at the corresponding position of the three-dimensional image, the method comprises: in response to an image zoom instruction, synchronously zooming the brain tissue segmentation image and the three-dimensional image, and superimposedly displaying the zoomed brain tissue segmentation image at a corresponding position of the zoomed three-dimensional image.

8. An electronic device, comprising: comprise: a processor and a memory for storing a computer program capable of running on the processor, wherein the processor is configured to execute the computer program to perform the steps of any one of claims 1 to 7.

9. A storage medium having stored thereon a computer program, characterized in that The computer program is executed by the processor to implement the steps of any one of claims 1 to 7.

10. A computer product comprising a computer program, characterized in that The computer program is executed by the processor to implement the steps of any one of claims 1 to 7. The computer program is executed by the processor to implement the steps of any one of claims 1 to 7.

Images