Image processing method and device, equipment, medium and program product

Abstract

The invention discloses an image processing method and device, equipment, a medium and a program product. The method comprises the following steps: acquiring three-dimensional image data and a two-dimensional projection image of a target bone segment; determining a target bone part in the two-dimensional projection image of the target bone segment according to the projection direction of the two-dimensional projection image of the target bone segment; segmenting the three-dimensional image data of the target bone segment to obtain three-dimensional image data of the target bone part; determining two-dimensional contour data of the target skeleton part based on the three-dimensional image data of the target skeleton part; and superposing the two-dimensional contour data of the target bone part into the two-dimensional projection image of the target bone section to generate a target image marked with the contour of the target bone part. According to the method, the display effect of the image is remarkably improved, and the readability and the information amount of the image in medical diagnosis and research are enhanced.

Description

Technical Field

[0001] This application belongs to the field of computer technology, and particularly relates to an image processing method, apparatus, device, medium, and program product. Background Art

[0002] In bone surgery, preoperative three-dimensional CT images can usually clearly present the bone anatomical structure, while intraoperative two-dimensional X-ray data reflects the bone characteristics of the patient's current position in real time. To achieve precise surgical navigation, the preoperative three-dimensional image can be registered with the intraoperative two-dimensional image. This registration process first requires converting the three-dimensional image into a two-dimensional image through perspective projection.

[0003] However, in the process of converting a three-dimensional image into a two-dimensional image, the prior art often faces many challenges, such as image detail loss, projection distortion, bone structure occlusion, etc., which affect the conversion effect and thus reduce the accuracy of surgical navigation.

[0004] Therefore, how to design a technical solution that can effectively improve the conversion effect from a three-dimensional image to a two-dimensional image has become an urgent technical problem to be solved. Summary of the Invention

[0005] Embodiments of this application provide an image processing method, apparatus, device, medium, and program product, which can improve the quality of the two-dimensional image obtained by converting a three-dimensional image.

[0006] In a first aspect, embodiments of this application provide an image processing method, which includes:

[0007] Obtain the three-dimensional image data and two-dimensional projection image of the target bone segment. The two-dimensional projection image of the target bone segment is obtained by performing projection processing on the three-dimensional image data of the target bone segment;

[0008] Determine the target bone part in the two-dimensional projection image of the target bone segment according to the projection direction of the two-dimensional projection image of the target bone segment;

[0009] Segment the three-dimensional image data of the target bone segment to obtain the three-dimensional image data of the target bone part;

[0010] Determine the two-dimensional contour data of the target bone part based on the three-dimensional image data of the target bone part;

[0011] Overlay the two-dimensional contour data of the target bone part onto the two-dimensional projection image of the target bone segment to generate a target image marked with the contour of the target bone part.

[0012] In a second aspect, embodiments of this application provide an image processing apparatus, which includes:

[0013] An acquisition module, configured to acquire three-dimensional image data and two-dimensional projection images of a target bone segment, where the two-dimensional projection image of the target bone segment is obtained by performing projection processing on the three-dimensional image data of the target bone segment;

[0014] An analysis module, configured to determine a target bone part in the two-dimensional projection image of the target bone segment according to the projection direction of the two-dimensional projection image of the target bone segment;

[0015] A segmentation module, configured to segment the three-dimensional image data of the target bone segment to obtain three-dimensional image data of the target bone part;

[0016] A processing module, configured to determine contour data of the target bone part based on the three-dimensional image data of the target bone part;

[0017] A synthesis module, configured to superimpose the contour data of the target bone part onto the two-dimensional projection image of the target bone segment to generate a target image marked with the contour of the target bone part.

[0018] In a third aspect, an embodiment of the present application provides a computer device, including: a processor, and a memory storing computer program instructions; the processor reads and executes the computer program instructions to implement the method according to any one of the first aspects above.

[0019] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the method according to any one of the first aspects above is implemented.

[0020] In a fifth aspect, an embodiment of the present application provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the method according to any one of the first aspects above is implemented.

[0021] The technical solutions provided by the embodiments of the present application at least bring the following beneficial effects:

[0022] The image processing method provided by the embodiment of the present application accurately segments the three-dimensional image data of the target bone part from the three-dimensional image data of the bone segment according to the two-dimensional projection direction of the bone segment; based on the segmented three-dimensional image data, extracts the two-dimensional contour information of the target bone part; combines the two-dimensional contour information of the target bone part with the two-dimensional image of the bone segment to generate a two-dimensional image with clear structural contours and prominent details. This method ensures that the extracted contour information is highly consistent with the actual anatomical morphology through three-dimensional image segmentation, and at the same time combines the contour information of three-dimensional data with the details of two-dimensional images to highlight key features, significantly improving the display effect of images and enhancing the readability and information content of images in medical diagnosis and research.

[0023] It should be understood that the above general description and the following detailed description are merely exemplary and explanatory, and do not limit this application. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The drawings herein are incorporated into the specification and constitute a part of this specification, showing embodiments consistent with this application, and are used together with the specification to explain the principles of this application, and do not constitute an undue limitation on this application.

[0025] Figure 1 is a schematic flowchart of an image processing method provided by an embodiment of this application;

[0026] Figure 2 is a schematic flowchart of an image processing method provided by an embodiment of this application;

[0027] Figure 3 is a schematic flowchart of an image processing method provided by an embodiment of this application;

[0028] Figure 4 is a schematic example diagram of each bone part in the vertebral segment provided by an embodiment of this application;

[0029] Figure 5 is a schematic example diagram of each bone part in the sacrum provided by an embodiment of this application;

[0030] Figure 6 is a schematic example diagram of the center line of the left pedicle channel provided by an embodiment of this application;

[0031] Figure 7 is a schematic example diagram of the center line of the right pedicle channel provided by an embodiment of this application;

[0032] Figure 8 is a schematic example diagram of obtaining binary feature data of the pedicle provided by an embodiment of this application;

[0033] Figure 9 is a schematic example diagram of obtaining binary feature data of the spinous process in the vertebral segment provided by an embodiment of this application;

[0034] Figure 10 is a schematic example diagram of obtaining binary feature data of the vertebral body in the vertebral segment provided by an embodiment of this application;

[0035] Figure 11 is a schematic example diagram of the sacral coordinate system provided by an embodiment of this application;

[0036] Figure 12 is a schematic example diagram of obtaining binary feature data of the left arc surface provided by an embodiment of this application;

[0037] Figure 13 is a schematic example diagram of obtaining binary feature data of the right arc surface provided by an embodiment of this application;

[0038] Figure 14 It is a schematic example diagram for obtaining the binarized feature data of the spinous process in the sacrum provided by an embodiment of the present application;

[0039] Figure 15 It is a schematic example diagram for obtaining the binarized feature data of the vertebral body in the sacrum provided by an embodiment of the present application;

[0040] Figure 16 It shows a schematic example diagram of the projection contours of each bone part in a vertebral segment provided by an embodiment of the present application;

[0041] Figure 17 It shows a schematic example diagram of the projection contours of each bone part in the sacrum provided by an embodiment of the present application;

[0042] Figure 18 It is a schematic structural diagram of an image processing device provided by an embodiment of the present application;

[0043] Figure 19 It is a schematic structural diagram of a computer device provided by an embodiment of the present application. Detailed implementation manners

[0044] In order to enable those of ordinary skill in the art to better understand the technical solutions of 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. It should be understood that the specific embodiments described herein are only intended to explain the present application, rather than limiting the present application. For those skilled in the art, the present application can be implemented without some of these specific details. The following description of the embodiments is only intended to provide a better understanding of the present application by showing examples of the present application.

[0045] It should be noted that the terms "first", "second", etc. in the specification, claims and above-mentioned drawings of the present application are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence. It should be understood that such used data can be interchanged under appropriate circumstances so that the embodiments of the present application described herein can be implemented in an order different from those illustrated or described herein. The implementation manners described in the following exemplary embodiments do not represent all implementation manners consistent with the present application. On the contrary, they are only examples consistent with some aspects of the present application as detailed in the appended claims.

[0046] As described in the background art section, the prior art lacks a method that can effectively retain the details of the bone structure during the conversion of three-dimensional images into two-dimensional images. To solve this technical problem, embodiments of the present application provide an image processing method, apparatus, device, medium, and program product, which extract the outlines of each part of the structure from the three-dimensional image data of the bone segment and combine them with the two-dimensional image of the bone segment to generate a two-dimensional image with clear structural outlines and prominent details, thereby improving the display effect of the image.

[0047] The following will combine the accompanying drawings to elaborate in detail on the image processing method provided by the embodiments of the present application through specific embodiments.

[0048] Figure 1 It is a schematic flowchart of an image processing method provided by an embodiment of the present application.

[0049] As Figure 1 shown, the image processing method provided by the embodiments of the present application may include step 110-step 150.

[0050] Step 110, obtain the three-dimensional image data and two-dimensional projection image of the target bone segment.

[0051] The target bone segment may be a specific bone part / region selected for processing or research in surgery or image analysis.

[0052] In the embodiments given in the present application, the target bone segment may be a vertebral segment, specifically including ordinary vertebral segments and / or the sacrum.

[0053] The three-dimensional image data may be medical image data of the target bone segment collected based on a medical imaging device, such as Computed Tomography (CT) image data.

[0054] In some embodiments of the present application, obtaining the three-dimensional image data of the target bone segment may include: performing segmentation processing on the medical image data of the target object to obtain the three-dimensional image data of the target bone segment.

[0055] Among them, the target object may be an object to undergo bone surgery, such as a patient to undergo spinal surgery.

[0056] In some embodiments of the present application, the two-dimensional projection image of the target bone segment may be obtained by projecting the three-dimensional image data.

[0057] It should be understood that according to different projection directions, the two-dimensional projection image of the target bone segment can be divided into a frontal two-dimensional projection image and a lateral two-dimensional projection image.

[0058] The anterior two-dimensional projection image is obtained by projecting three-dimensional imaging data in the frontal direction of the target object, and can provide the bone structure observed from the frontal perspective of the target object.

[0059] The lateral two-dimensional projection image is obtained by projecting three-dimensional imaging data in the lateral direction corresponding to the target, and can provide the bone structure observed from the lateral perspective of the target object.

[0060] Step 120: Determine the target bone part in the two-dimensional projection image of the target bone segment according to the projection direction of the two-dimensional projection image of the target bone segment.

[0061] The bone part can refer to a specific anatomical structure or part within the target bone segment.

[0062] The target bone part can refer to the bone part that can be observed and recognized in the projection direction of the two-dimensional projection image of the target bone segment.

[0063] It can be understood that since the two-dimensional projection image is taken according to the relative position between the radiation and the bone segment, different projection directions (such as anterior and lateral) will result in different views of the bone segment. For example, some bone parts may be blocked by other bones during anterior shooting, while they may be shown more clearly during lateral shooting.

[0064] Therefore, it is necessary to determine the bone part included in the two-dimensional projection image according to the projection direction of the two-dimensional projection image of the target bone segment.

[0065] In some embodiments of the present application, a mapping relationship between the projection direction and the type of bone part is preset. When obtaining the two-dimensional projection image of the target bone segment, determine its projection direction, and then the type of bone part included in the two-dimensional projection image in this projection direction can be queried based on this mapping relationship.

[0066] In some embodiments of the present application, a trained image recognition model is used to identify the type of bone part in the two-dimensional projection image.

[0067] Figure 4 Shows a schematic example diagram of each bone part in the vertebral segment. Exemplarily, as Figure 4 shown, when the target bone segment is the vertebral segment, the bone parts therein may include the spinous process, the upper surface of the vertebral body, the vertebral body, the left and right pedicles, the lower surface of the vertebral body, etc. When the projection direction is anterior, the bone parts corresponding to the two-dimensional projection image of the vertebral segment are determined to be the spinous process and the left and right pedicles. When the projection direction is lateral, the bone parts corresponding to the two-dimensional projection image of the vertebral segment are determined to be the upper edge of the vertebral body, the lower edge of the vertebral body, and the side of the vertebral body.

[0068] Figure 5 Shows a schematic example diagram of each bone part in the sacrum. Exemplarily, as Figure 5As shown, when the target bone segment is the sacrum, the bone parts therein may include the spinous process and the vertebral body (superior vertebral surface, left arc surface of the vertebral body, right arc surface of the vertebral body). When the projection direction is the anteroposterior view, the bone parts corresponding to the two-dimensional projection image of the sacrum are determined as the spinous process and the superior vertebral surface. When the projection direction is the lateral view, the bone parts corresponding to the two-dimensional projection image of the sacrum are determined as the lateral side of the sacrum and the sacral projection semi-circle (formed by the left arc surface and the right arc surface of the vertebral body).

[0069] Step 120: Segment the three-dimensional image data of the target bone segment to obtain the three-dimensional image data of the target bone part.

[0070] In one implementation, the three-dimensional image data of each bone part can be segmented from the three-dimensional image data of the target bone segment. This implementation is applicable to scenarios where a comprehensive understanding of the anatomical structure of the entire vertebral segment is required.

[0071] In one implementation, the three-dimensional image data of the target bone part can be segmented from the three-dimensional image data of the target bone segment. The target vertebral structure can be clearly specified according to the preoperative plan or the key structure selected according to the lesion area, and can be one or more. This implementation focuses on specific vertebral parts, can more efficiently obtain information on the target area, and can reduce the computational burden of processing and storing unnecessary data.

[0072] Step 130: Based on the three-dimensional image data of the target bone part, determine the two-dimensional contour data of the target bone part.

[0073] The two-dimensional contour data can refer to the boundary information of the target bone part in a specific projection direction or slice plane, and is used to describe the two-dimensional representation of the shape, size and position of the bone part.

[0074] Exemplarily, the two-dimensional contour data can be composed of a series of points, lines or closed curves, and can be represented by a mathematical model (such as a polygon, a spline curve) or pixel data.

[0075] In some embodiments of the present application, the three-dimensional image data of the target bone part is converted into a two-dimensional image, and an edge detection algorithm is used on the two-dimensional image to extract the two-dimensional contour data of the target bone part. This method is fast and simple, and is applicable to scenarios where contour data in a single projection direction is required.

[0076] In some embodiments of the present application, multiple slices parallel to the selected plane are extracted from the three-dimensional image data of the target bone part, the bone regions of each slice are segmented, a contour extraction algorithm is applied to obtain the contour data of each slice. The slice contours are superimposed according to the projection direction and fused into a complete two-dimensional contour map. This method is more refined, can adapt to the contour fusion of multiple projection planes, and is applicable to the analysis requirements of complex bone parts.

[0077] Step 140: Overlay the two-dimensional contour data of the target bone part onto the two-dimensional projection image of the target bone segment to generate a target image marked with the vertebral structure contour.

[0078] This step is to overlay the two-dimensional contour data extracted from the target bone part onto the two-dimensional projection image of the corresponding target bone segment to generate a combined image. This combined image can intuitively display the original image details of the target bone segment, while the overlaid contour data provides clear boundary and morphological information of the bone part, thus forming a target image marked with the contour of the vertebral part.

[0079] In some embodiments of the present application, according to the coordinate systems of the contour data and the projection image, necessary registration and alignment operations are performed to ensure the matching of their scales and positions. The contour data is converted into pixel form, and contour lines are drawn on the projection image. Exemplarily, different colors can be used to mark the contour for highlighting. Finally, a combined image containing the original projection information and the overlaid contour is output. This method is simple and easy to implement and is suitable for scenarios where data alignment is relatively simple.

[0080] In some embodiments of the present application, the two-dimensional contour of the target bone is extracted from the three-dimensional image data and projected onto a plane consistent with the target two-dimensional projection image. The contour data is fused with the original projection image in a semi-transparent or stroked manner. The visual attributes such as the transparency, color, and thickness of the contour are adjusted to make it clear on the original image without obscuring details. A target image with an overlaid effect is generated, and at the same time, user interaction is supported to adjust the overlaid effect (such as rotation, scaling). This method has a natural fusion effect and a better visual experience and is suitable for more complex image data scenarios.

[0081] The image processing method provided by the embodiments of the present application accurately segments the three-dimensional image data of each part structure from the three-dimensional image data of the bone segment according to the two-dimensional projection direction of the bone segment; based on the segmented three-dimensional image data, extracts the two-dimensional contour information of each part structure; combines the two-dimensional contour information of each part structure with the two-dimensional image of the bone segment to generate a two-dimensional image with clear structure contours and prominent details. This method significantly improves the display effect of the image through the fusion processing of three-dimensional data and two-dimensional images, enhancing the readability and information content of the images in medical diagnosis and research.

[0082] Figure 2 The flowchart of an image processing method provided by the embodiments of the present application is shown.

[0083] It should be understood that Figure 2 The illustrated embodiment can be regarded as an example of step 130.

[0084] Such as Figure 2As shown, the method may include the following steps.

[0085] Step 210: Determine the regional range corresponding to the target bone part in the three-dimensional image data of the target bone segment.

[0086] In one implementation, the actual regional range of the target bone can be determined according to the anatomical characteristics of the target bone segment by using a geometric model or a mathematical model.

[0087] Exemplarily, the geometric model or the mathematical model is constructed based on the prior knowledge of the structure of the target bone segment.

[0088] Step 220: Generate the binary feature data of the target bone part according to the regional range corresponding to the target bone part.

[0089] The binary feature data may refer to three-dimensional data in which the area where the target bone part is located is represented by "1" and other areas are represented by "0", and is used to extract the image information of the target bone part from the three-dimensional image data.

[0090] Exemplarily, the binary feature data may be NEED (Normalized Enhanced Edge Detection) data.

[0091] In one implementation, according to the located regional range of the target bone part, the corresponding voxels are assigned the value "1" and the background is assigned the value "0", and the generated binary feature data can be saved in a specific format, such as the standard medical image data format or a custom format (such as NEED data).

[0092] Step 230: Use the binary feature data of the target bone part to segment the three-dimensional image data of the target bone part from the three-dimensional image data of the target bone segment.

[0093] Specifically, the binary feature data is multiplied by the three-dimensional image data of the target bone segment voxel by voxel, and only the image information of the target bone part is retained. The binary feature data is used as a mask (Mask) to directly extract the target bone part area from the three-dimensional image data.

[0094] In this embodiment, through the precise positioning of the target bone part area, the binary processing of the feature data, and the three-dimensional image segmentation based on the binary features, the three-dimensional image data of the target bone part is efficiently and accurately extracted, which is convenient for the accurate recognition of the subsequent contour.

[0095] In one embodiment, when the target bone segment is a vertebral segment, the target bone part includes the pedicle. To segment the three-dimensional image data of the pedicle from the three-dimensional image data of the vertebral segment, the following steps may be included.

[0096] Step 211: Calculate the center line of the pedicle channel based on the characteristics of the vertebral segment.

[0097] Step 212: Based on the center line of the pedicle channel, respectively locate the area range corresponding to the left pedicle and the area range corresponding to the right pedicle in the three-dimensional image data of the vertebral segment.

[0098] It should be understood that Step 211 and Step 212 can be regarded as an example of Step S210.

[0099] Step 221: Generate first binary feature data according to the area range corresponding to the left pedicle.

[0100] Step 222: Generate second binary feature data according to the area range corresponding to the right pedicle.

[0101] Step 223: Based on the Boolean algorithm, merge the first binary feature data and the second binary feature data to obtain the binary feature data of the pedicle.

[0102] It should be understood that Step 221 - Step 223 can be regarded as an example of Step S220.

[0103] In this embodiment, by mathematically modeling the center line of the pedicle channel, locating the area range of the pedicle in the three-dimensional image data of the vertebral segment, and combining the Boolean algorithm to extract the binary feature data, the accurate segmentation of the pedicle is realized. This method utilizes the geometric characteristics of the channel center line, improves the accuracy and efficiency of segmentation, and lays a reliable foundation for the subsequent analysis and processing of the bone part.

[0104] Figures 6 - 8 For a schematic example of obtaining the binary feature data of the pedicle, the following is combined with Figures 6 - 8 to introduce an example of obtaining the binary feature data of the pedicle.

[0105] Calculate the center lines of the left and right pedicle channels through the characteristics of the vertebral segment. The center line is at the exact center of the pedicle channel, and the midpoint of the center line is at the exact center of the isthmus of the pedicle channel.

[0106] The center line of the left pedicle channel is as shown in Figure 6 and the center line of the right pedicle channel is as shown in Figure 7 as shown.

[0107] As shown in Figure 8 a cylinder is generated along the center line of the left pedicle channel. The parameters of the cylinder can be set as: radius 15, height 20, and extending 10 from the midpoint of the center line to both ends. The center line of the cylinder is the center line of the left pedicle channel. Extract the vertebral segment data inside the cylinder to obtain the NRRD data of the left pedicle.

[0108] It should be understood that the area inside the cylinder is the area range corresponding to the left pedicle.

[0109] Extract the vertebral segment data inside the cylinder to obtain the NRRD data of the left pedicle.

[0110] Exemplarily, when extracting the vertebral segment data inside the cylinder, the cylinder can be converted into VTK image data (VTK Image Data) type using polygon data conversion to image template (VTK PolyData To Image Stencil), and then converted into ITK image (ITK Image) type, and a Boolean operation is performed with the vertebral segment NRRD data to obtain the valid data of the left pedicle.

[0111] Similarly, as Figure 8 shown, a cylinder is generated along the center line of the right pedicle channel. The parameters of the cylinder can be set as: radius 15, height 20. Similarly, with the midpoint of the center line as the reference, it extends 10 at both ends, and the center line is the center line of the right pedicle channel.

[0112] It should be understood that the area inside the cylinder is the area range corresponding to the right pedicle.

[0113] Extract the vertebral segment data inside the cylinder to obtain the NRRD data of the right pedicle.

[0114] Finally, the NRRD data of the left and right pedicles are merged through Boolean operation to generate the complete NRRD data of the pedicle.

[0115] In one embodiment, when the target bone segment is a vertebral segment, the target bone part includes the spinous process. To segment the three-dimensional image data of the spinous process from the three-dimensional image data of the vertebral segment, the following steps may be included.

[0116] Construct multiple planes based on the center line of the pedicle channel, and locate the area range corresponding to the spinous process in the three-dimensional image data of the target bone segment according to the multiple planes and preset displacement parameters.

[0117] Generate the binary feature data of the spinous process according to the area range corresponding to the spinous process.

[0118] Use the binary feature data of the spinous process to segment the three-dimensional image data of the spinous process from the three-dimensional image data of the vertebral segment.

[0119] In this embodiment, by constructing planes based on the center line of the pedicle channel and combining preset displacement parameters to locate the area range of the spinous process and generate its binary feature data, the three-dimensional image data of the spinous process is accurately segmented.

[0120] Figure 9 For a schematic example diagram of obtaining the binary feature data of the spinous process, the following is combined withFigure 9 Introduce an example of obtaining binarized feature data of the spinous process.

[0121] Taking the midpoint of the line connecting the midpoints of the centerlines of the left and right pedicle channels as the reference, a plane is generated, and the normal vector of this plane is (0, 1, 0). Subsequently, with this plane as the reference, it is translated parallel to the x-axis by -7.5 units to generate a new plane with a normal vector of (1, 0, 0). Then, it is translated parallel to the x-axis by +7.5 units again to generate another plane with a normal vector of (-1, 0, 0).

[0122] Using the above three planes to crop the vertebral segment data, the enclosed area is the area range corresponding to the spinous process.

[0123] Extract the data within the enclosed area to obtain the NRRD data of the spinous process.

[0124] In one embodiment, when the target bone segment is a vertebral segment, the target bone part includes the vertebral body. To segment the three-dimensional image data of the vertebral body from the three-dimensional image data of the vertebral segment, the following steps may be included.

[0125] Based on the centerline of the pedicle channel, the area range corresponding to the spinous process, and the area range corresponding to the pedicle, locate the area range corresponding to the vertebral body in the three-dimensional image data of the vertebral segment.

[0126] Generate the binarized feature data of the vertebral body according to the area range corresponding to the vertebral body.

[0127] Use the binarized feature data of the vertebral body to segment the three-dimensional image data of the vertebral body from the three-dimensional image data of the vertebral segment.

[0128] In this embodiment, by combining the spatial features of the centerline of the pedicle channel, the spinous process, and the area range corresponding to the pedicle, the area range of the vertebral body is accurately located in the three-dimensional image data of the vertebral segment, the binarized feature data of the vertebral body is generated, and the three-dimensional image data of the vertebral body is segmented. This method utilizes the spatial correlation of the key anatomical features of the target bone segment to achieve the accurate extraction of the vertebral body area.

[0129] Figure 10 For a schematic example diagram of obtaining the binarized feature data of the vertebral body, the following is combined with Figure 10 Introduce an example of obtaining the binarized feature data of the vertebral body.

[0130] Taking the midpoint of the centerlines of the left and right pedicle channels as the reference, a plane perpendicular to the centerline of the pedicle is generated, and its normal vector is the direction vector from the centerline of the pedicle to the vertebral body. This plane is used to segment the vertebral segment data.

[0131] Use these two planes to segment the vertebral segment NRRD data, removing the pedicle and spinous process parts. The remaining vertebral segment data after segmentation is the NRRD data of the vertebral body.

[0132] Search downward in the vertebral body NRRD along the Z-axis of the three-dimensional image data of the vertebral segment. The set of the first non-zero points found is the upper surface NRRD data.

[0133] Search upward in the vertebral body NRRD along the Z-axis of the three-dimensional image data of the vertebral segment. The set of the first non-zero points found is the lower surface NRRD data.

[0134] In one embodiment, when the target bone segment is the sacrum, the target bone part includes the left arc surface and the right arc surface. Segmenting the three-dimensional image data of the left arc surface and the right arc surface from the three-dimensional image data of the sacrum may include the following steps.

[0135] Calculate the sacral coordinate system based on the characteristics of the sacrum.

[0136] Create a first cube based on the sacral coordinate system and the left sacral foramen position, and determine the first cube as the regional range corresponding to the left arc surface.

[0137] Create a second cube based on the sacral coordinate system and the right sacral foramen position, and determine the second cube as the regional range corresponding to the right arc surface.

[0138] Generate the binary feature data of the left arc surface according to the regional range corresponding to the left arc surface.

[0139] Generate the binary feature data of the right arc surface according to the regional range corresponding to the right arc surface.

[0140] Use the binary feature data of the left arc surface and the right arc surface to segment the three-dimensional image data of the left arc surface and the right arc surface from the three-dimensional image data of the sacrum.

[0141] In this embodiment, by calculating the sacral coordinate system and combining the position characteristics of the left and right sacral foramina, the corresponding cube regional ranges are created respectively, and the binary feature data of the left arc surface and the right arc surface are generated, so as to accurately segment the three-dimensional image data of the left arc surface and the right arc surface from the three-dimensional image data of the sacrum. This method utilizes the matching of the sacral anatomical characteristics and the coordinate system to achieve the efficient positioning and segmentation of the left and right arc surfaces.

[0142] Figure 11 It is a schematic example diagram of the sacral coordinate system.

[0143] Figure 12 It is a schematic example diagram for obtaining the binary feature data of the left arc surface. The following combines Figure 12 Introduce an example of obtaining the binary feature data of the left arc surface.

[0144] Calculate the sacral coordinate system based on the characteristics of the sacrum, and determine the positions of the first pair of sacral foramina.

[0145] In Figure 11 the yellow point is the origin of the sacral coordinate system, the red ray to the left is the positive X-axis, the green ray forward is the positive Y-axis, and the blue ray upward is the positive Z-axis. The red dot is marked as the left first sacral foramen, and the blue dot is marked as the right first sacral foramen.

[0146] As Figure 12 shown, rotate counterclockwise by 20 degrees around the Y-axis of the sacral coordinate system to generate a new coordinate system.

[0147] Taking the position of the left first sacral foramen as the center of the bottom surface, generate a cuboid in the new coordinate system.

[0148] The parameters of the cuboid can be set as: length 2.5, width 20, and height 40. The edges of the cuboid are parallel to the X, Y, and Z axes of the new coordinate system.

[0149] Use the VTK PolyData To Image Stencil class to convert the cuboid into VTK Image Data type, then convert it into ITK Image type, and perform a Boolean operation with the vertebral segment NRRD data to extract the vertebral segment data within the cuboid and generate the left arc surface NRRD data.

[0150] Figure 13 For an example diagram of obtaining the binary feature data of the right arc surface, the following is an example of obtaining the binary feature data of the right arc surface in combination with Figure 13 this.

[0151] Rotate clockwise by 20 degrees around the Y-axis of the sacral coordinate system to generate a new coordinate system.

[0152] Taking the position of the right first sacral foramen as the center of the bottom surface, generate a cuboid in the new coordinate system:

[0153] The parameters of the cuboid can be set as: length 2.5, width 20, and height 40.

[0154] The edges of the cuboid are parallel to the X, Y, and Z axes of the new coordinate system.

[0155] According to the method of obtaining the left arc surface NRRD data, also use VTK PolyData To Image Stencil and Boolean operation to extract the vertebral segment data within the cuboid and generate the right arc surface NRRD data.

[0156] Through the above methods, the binary feature data of the left arc surface and the right arc surface are obtained respectively, thus realizing the accurate extraction of the bone characteristics in the sacral region.

[0157] In one embodiment, when the target bone segment is the sacrum, the target bone site includes the sacral spinous process and the sacral vertebral body. The three-dimensional image data of the spinous process and the vertebral body can be segmented from the three-dimensional image data of the sacrum, which may include the following steps.

[0158] Construct multiple planes based on the sacral coordinate system. Based on the multiple planes, locate the corresponding regional ranges of the sacral spinous process and the sacral vertebral body in the three-dimensional image data of the sacrum.

[0159] Generate the binary feature data of the sacral spinous process according to the corresponding regional range of the sacral spinous process.

[0160] Generate the binary feature data of the sacral vertebral body according to the corresponding regional range of the sacral vertebral body.

[0161] Use the binary feature data of the sacral spinous process to segment the three-dimensional image data of the sacral spinous process from the three-dimensional image data of the sacrum.

[0162] Use the binary feature data of the sacral vertebral body to segment the three-dimensional image data of the sacral vertebral body from the three-dimensional image data of the sacrum.

[0163] In this embodiment, multiple planes are constructed based on the sacral coordinate system. Through these planes, the corresponding regional ranges of the spinous process and the vertebral body are accurately located in the three-dimensional image data of the sacrum. The binary feature data of the spinous process and the vertebral body are respectively generated, and finally the three-dimensional image data of the spinous process and the vertebral body are segmented from the three-dimensional image data of the sacrum.

[0164] Figure 14 For a schematic example diagram of obtaining the binary feature data of the spinous process in the sacrum, the following is combined with Figure 14 Introduce an example of obtaining the binary feature data of the spinous process in the sacrum.

[0165] Taking the origin of the sacral coordinate system as the center, with the +Y axis as the normal vector, rotate 30 degrees counterclockwise around the Z axis to generate a plane; then taking the origin of the sacral coordinate system as the center, with the +Y axis as the normal vector, rotate 30 degrees clockwise around the Z axis to generate a plane; then taking the origin of the sacral coordinate system as the center and the +Y axis of the coordinate system as the normal vector, generate a plane. Take the vertebral segment data between the three planes, which is the NRRD data of the spinous process.

[0166] Figure 15 For a schematic example diagram of obtaining the binary feature data of the vertebral body in the sacrum, the following is combined with Figure 15 Introduce an example of obtaining the binary feature data of the vertebral body in the sacrum.

[0167] Starting from the origin of the sacral coordinate system, move (0, -10, -5) along the coordinate directions to obtain a new point P. Taking the Z-axis of the sacral coordinate system as the central axis, with point P as the center of the cylinder base, a radius of 25, and a height of 30 (extending 15 in both the positive and negative Z-axis directions), generate a cylinder. Use the cylinder as a mask to crop the vertebral segment data, and the data inside the cylinder is the upper vertebral surface NRRD data.

[0168] Perform a Boolean operation on the sacral NRRD data and the spinous process NRRD data to remove the spinous process part and obtain the sacral data after removing the spinous process.

[0169] Perform a Boolean operation on the sacral data after removing the spinous process and the upper vertebral surface NRRD data to remove the upper vertebral surface part and obtain the further segmented sacral data.

[0170] Starting from the origin of the sacral coordinate system, move 40 along the positive X-axis direction to generate a plane, and the normal vector of the plane is the -X-axis.

[0171] Starting from the origin of the sacral coordinate system, move 40 along the negative X-axis direction to generate another plane, and the normal vector of the plane is the X-axis.

[0172] Use the above two planes to crop the sacral data, extract the vertebral segment data between the two planes, and obtain the vertebral body NRRD data.

[0173] Through the above steps, the binary feature data of the vertebral body in the sacrum can be obtained, and the spinous process and the upper vertebral surface part are ensured to be removed, making the result data more accurate.

[0174] Figure 3 The flowchart of an image processing method provided by an embodiment of the present application is shown.

[0175] It should be understood that Figure 3 The shown embodiment can be regarded as an example of step 140.

[0176] Such as Figure 3 As shown, the method may include the following steps.

[0177] Step 310: Perform projection processing on the three-dimensional image data of the target bone part to obtain a two-dimensional projection image of the target bone part.

[0178] Among them, the projection direction of the two-dimensional projection image of the target bone part is consistent with the projection direction of the two-dimensional projection image of the target bone segment.

[0179] Step 320: Perform binarization processing on the two-dimensional projection image of the target bone part to obtain a binary image of the target bone part.

[0180] Exemplarily, the binary image can be a Digitally Reconstructed Radiograph (DRR).

[0181] Step 330: Identify the contours in the binary image, determine the target contours of the target bone part that meet the preset conditions, and obtain the contour data of the target contours.

[0182] In this embodiment, by performing projection processing on the three-dimensional image data of the target bone part, a two-dimensional projection image consistent with the projection direction of the target bone segment is generated, and the contours are identified after binarization processing. Finally, the target contours and their contour data are extracted according to the preset conditions. The two-dimensional contour data of the target bone part obtained by this method is based on the three-dimensional image data, so it can match the two-dimensional projection image of the target bone segment, thereby improving the accurate recognition of the bone part and the accuracy of subsequent analysis.

[0183] In one embodiment, when the target bone segment is a vertebral segment and the projection direction is a frontal projection, the target bone part includes the spinous process and / or the pedicle. Identifying the contours in the binary image and determining the target contours of the target bone part that meet the preset conditions may include the following steps.

[0184] Identify all the contours in the binary image corresponding to the spinous process, and determine the target contour of the spinous process as the contour with the largest area among all the contours.

[0185] In one example, call the OpenCV Threshold Function in the OpenCV library to perform binarization processing on the DRR image of the spinous process, and then call the OpenCV FindContours Function to extract all the contours in the binarized image, and find the contour with the largest area among these contours and define it as the spinous process contour.

[0186] Identify all the contours in the binary image corresponding to the pedicle, screen out the contours of the left pedicle and the right pedicle according to the area size, perform pixel brightness search within a preset radius based on the centroids of the contours of the left pedicle and the right pedicle, and determine the target contour of the pedicle according to the set of pixel points that reach the preset brightness searched out.

[0187] In one example, after binarizing the DRR images of the left and right pedicles, all contours are extracted and their areas are calculated. The contour with the largest area is selected and named contours1. Then, for the other contours, the distances from all their points' abscissas to contours1 are calculated, and the contours with distances greater than 80 pixels are screened out. Among them, the contour with the largest area is selected and named contours2. Then, the average abscissas of contours1 and contours2 are calculated respectively. The contour with the smaller average abscissa is the left pedicle contour, and the larger one is the right pedicle contour. Further, the centroids of the left and right pedicle contours are calculated, and with the centroid as the origin, points with the maximum brightness are searched within a certain range (10 to 60 pixels) in the polar coordinate radius direction, and the set of these points is determined as the pedicle contour.

[0188] In one embodiment, when the target bone segment is a vertebral segment and the projection direction is a lateral projection, the target bone part includes the vertebral body. Identifying the contours in the binary image and determining the target contours of the target bone part that meet the preset conditions may include the following steps.

[0189] Identify all the contours in the binary image corresponding to the vertebral body, and screen out the contour with the largest area among all the contours;

[0190] Obtain the two-dimensional projection image of the region corresponding to the contour with the largest area;

[0191] Determine the target contour of the vertebral body according to the set of pixel points that reach the preset brightness in the preset direction in the two-dimensional projection image of the region.

[0192] In one example, the target contour of the vertebral body may be composed of the upper contour of the vertebral body, the lower contour of the vertebral body, and the side contour of the vertebral body.

[0193] Upper contour of the vertebral body: Binarize the lateral DRR projection image of the upper surface of the vertebral body, extract all contours and find the contour with the largest area, generate a mask with it, cut out the corresponding image area from the lateral DRR projection image of the vertebral body, and then search for the maximum brightness within a certain distance downward along the Y-axis direction of the image. The set of obtained points is determined as the upper contour of the vertebral body.

[0194] Lower contour of the vertebral body: Binarize the lateral DRR projection image of the lower surface of the vertebral body, extract all contours and find the contour with the largest area, generate a mask with it, cut out the corresponding image area from the lateral DRR projection image of the vertebral body, and then search for the maximum brightness within a certain distance upward along the negative Y-axis direction of the image. The set of obtained points is determined as the lower contour of the vertebral body.

[0195] Lateral side of the vertebral body: Binarize the lateral DRR projection image of the vertebral body, extract all contours and calculate the indices of the four corner points, and crop the left and right side regions with these corner points; when the bed is on the left, take the right side as the contour of the lateral side of the vertebral body, and when the bed is on the right, take the left side as the contour of the lateral side of the vertebral body.

[0196] In one embodiment, when the target bone segment is the sacrum and the projection direction is the anteroposterior projection, the target bone part includes the upper edge of the vertebral body and / or the sacral spinous process. Identify the contours in the binary image, and determine the target contours of the target bone part that meet the preset conditions, which may include the following steps.

[0197] Identify all the contours in the binary image corresponding to the upper edge of the vertebral body, select the contour with the largest area among all the contours, segment the contour with the largest area, and determine the segmented contour as the target contour of the upper edge of the vertebral body.

[0198] In one example, binarize the lateral DRR projection image of the upper surface of the vertebral body, extract all contours and find the contour with the largest area; then calculate the coordinates of the leftmost point and the rightmost point in this contour, and use the coordinates of these two points to segment the lower half of the contour with the largest area, and the obtained part is determined as the anteroposterior contour of the sacral spinous process.

[0199] Skeletonize the binary image corresponding to the sacral spinous process to obtain the target contour of the sacral spinous process.

[0200] In one example, binarize the lateral DRR projection image of the upper surface of the vertebral body, and then call the Image Thinning Function (thinImage) function to thin the binary image to generate a polyline, and this polyline is determined as the anteroposterior contour of the spinous process.

[0201] In one embodiment, when the target bone segment is the sacrum and the projection direction is the lateral projection, the target bone part includes the lateral side of the sacrum and / or the arc surface of the sacrum. Identify the contours in the binary image, and determine the target contours of the target bone part that meet the preset conditions, which may include the following steps.

[0202] Identify all the contours in the binary image corresponding to the lateral side of the sacrum, select the contour with the largest area among all the contours, calculate the convex hull of the contour with the largest area, and based on the longest side of the convex hull, determine the set of pixel points with a preset brightness inside the binary image as the target contour of the lateral side of the sacrum.

[0203] In one example, the lateral DRR projection image of the vertebral body is binarized, all contours are extracted, and the contour with the largest area is found. The convex hull calculation function (Convex Hull Function) is called to calculate the convex hull of the contour with the largest area. Then, the lengths of the sides of the convex hull are calculated, the longest side and the second-longest side are found, and the distances from these two sides to the four vertices of the image are calculated to determine the side closer to the outside. Searching inward along this outside side, the points with the maximum brightness value are found, and the set of these points is the contour of the lateral side of the sacrum.

[0204] Identify the contours in the binary image corresponding to the sacral arc surface, filter out the contour with the largest area among all the contours, perform pixel brightness search within a preset radius range based on the centroid of the contour with the largest area, and determine the target contour of the sacral arc surface according to the set of pixel points that reach the preset brightness.

[0205] In one example, the DRR projection image of the left or right arc surface of the vertebral body (if the tail of the lateral sacral projection is on the left, the left arc surface is taken; otherwise, the right arc surface is taken) is binarized, all contours are extracted, and the contour with the largest area is found. The moment M of the contour with the largest area is calculated using the Moments Calculation Function, and the centroid coordinates of the contour are calculated through the moment formula: (x, y) = (M.m10 / M.m00, M.m01 / M.m00). Then, starting from this centroid, the points with the maximum brightness are searched within the polar coordinate range (20 - 80 pixels), and the set of these points is the contour of the sacral arc surface.

[0206] Figure 16 The schematic example diagram shows the projection contours of each bone part in the vertebral segment provided by the embodiment of the present application.

[0207] Figure 17 The schematic example diagram shows the projection contours of each bone part in the sacrum provided by the embodiment of the present application.

[0208] The above mainly introduced an image processing method according to an embodiment of the present application in combination with the accompanying drawings. At the same time, it should be understood that although the steps in the flowcharts involved in the above-mentioned embodiments are sequentially shown, these steps are not necessarily executed in the order shown in the figures. Unless there is a clear description in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least a part of other steps or steps in other steps. Next, an image processing device according to an embodiment of the present application will be introduced in combination with the accompanying drawings. For the sake of brevity, when introducing the device below, appropriate omissions will be made, and the relevant content can refer to the relevant descriptions in the above method and will not be repeated.

[0209] Corresponding to the method described in the above embodiment, Figure 18 FIG. shows a structural block diagram of an image processing device 1000 provided by an embodiment of the present application. For the sake of convenience of description, only the parts related to the embodiment of the present application are shown.

[0210] As Figure 18 shown, the device 1000 may include:

[0211] An acquisition module 1001, configured to acquire three-dimensional image data and two-dimensional projection images of a target bone segment.

[0212] The two-dimensional projection image of the target bone segment is obtained by performing projection processing on the three-dimensional image data of the target bone segment.

[0213] An analysis module 1002, configured to determine a target bone part in the two-dimensional projection image of the target bone segment according to the projection direction of the two-dimensional projection image of the target bone segment.

[0214] A segmentation module 1003, configured to segment the three-dimensional image data of the target bone segment to obtain three-dimensional image data of the target bone part.

[0215] A processing module 1004, configured to determine contour data of the target bone part based on the three-dimensional image data of the target bone part.

[0216] A synthesis module 1005, configured to superimpose the contour data of the target bone part onto the two-dimensional projection image of the target bone segment to generate a target image marked with the contour of the target bone part.

[0217] Figure 19 FIG. shows a schematic hardware structure diagram of a computer device provided by an embodiment of the present application.

[0218] The computer device may include a processor 7001 and a memory 7002 storing computer program instructions.

[0219] Specifically, the above-mentioned processor 7001 may include a Central Processing Unit (CPU), or an Application Specific Integrated Circuit (ASIC), or may be configured as one or more integrated circuits implementing the embodiments of the present application.

[0220] The memory 7002 may include a mass storage for data or instructions. By way of example and not limitation, the memory 7002 may include a Hard Disk Drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. In one example, the memory 7002 may include removable or non-removable (or fixed) media, or the memory 7002 is a non-volatile solid-state memory. The memory 7002 may be inside or outside the integrated gateway disaster recovery device.

[0221] In one example, the memory 7002 may be a Read Only Memory (ROM). In one example, the ROM may be a mask-programmed ROM, a Programmable ROM (PROM), an Erasable PROM (EPROM), an Electrically Erasable PROM (EEPROM), an Electrically Alterable ROM (EAROM), or a flash memory, or a combination of two or more of these.

[0222] The memory 302 may include a Read Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk storage media device, an optical storage media device, a flash memory device, an electrical, optical, or other physical / tangible memory storage device. Thus, generally, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described in reference to the method according to one aspect of the present disclosure.

[0223] The processor 7001 reads and executes the computer program instructions stored in the memory 7002 to implement Figure 1 the image processing method in the illustrated embodiment.

[0224] In one example, the computer device may further include a communication interface 7003 and a bus 7004. Among them, as Figure 19 shown, the processor 7001, the memory 7002, and the communication interface 7003 are connected through the bus 7004 and communicate with each other.

[0225] The communication interface 7003 is mainly used to implement communication between various modules, devices, units, and / or devices in the embodiments of the present application.

[0226] The bus 7004 includes hardware, software, or both, and couples the components of the online data flow meter charging device to each other. By way of example and not limitation, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a MicroChannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses or a combination of two or more of these. In a suitable case, the bus 7004 may include one or more buses. Although the embodiments of the present application describe and illustrate specific buses, the present application contemplates any suitable bus or interconnect.

[0227] In addition, in combination with the image processing method in the above embodiments, the embodiments of the present application may provide a computer storage medium to implement. Computer program instructions are stored on the computer storage medium; when the computer program instructions are executed by the processor, any one of the image processing methods in the above embodiments is implemented.

[0228] The embodiments of the present application also provide a computer program product, including a computer program, and when the computer program is executed by the processor, any one of the image processing methods in the above embodiments is implemented.

[0229] It should be clear that the present application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order between steps after understanding the spirit of the present application.

[0230] The functional blocks shown in the above-described structural block diagrams can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a functional card, and so on. When implemented in software, the elements of the present application are programs or code segments for performing the required tasks. The program or code segment can be stored in a machine-readable medium or transmitted via a data signal carried in a carrier wave on a transmission medium or a communication link. A "machine-readable medium" can include any medium that can store or transmit information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, read-only memory (ROM), flash memory, erasable read-only memory (EROM), floppy disks, compact disc read-only memory (CD-ROM), optical discs, hard disks, fiber optic media, radio frequency (RF) links, and so on. The code segment can be downloaded via a computer network such as the Internet, an intranet, and so on.

[0231] Aspects of the present disclosure have been described above with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present disclosure. It should be understood that each block in the flowcharts and / or block diagrams, and the combinations of blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing device enable the implementation of the functions / actions specified in one or more blocks of the flowcharts and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and the combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by dedicated hardware for performing the specified functions or actions, or by a combination of dedicated hardware and computer instructions.

[0232] As described above, this is only the specific implementation manner of the present application. Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the systems, modules, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein. It should be understood that the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of various equivalent modifications or substitutions, and these modifications or substitutions should all be covered within the protection scope of the present application.

Claims

1. An image processing method, characterized in that, Including: Obtaining three-dimensional image data and two-dimensional projection images of a target bone segment, where the two-dimensional projection image of the target bone segment is obtained by performing projection processing on the three-dimensional image data of the target bone segment; Determining a target bone part in the two-dimensional projection image of the target bone segment according to the projection direction of the two-dimensional projection image of the target bone segment; Segmenting the three-dimensional image data of the target bone segment to obtain three-dimensional image data of the target bone part; Determining two-dimensional contour data of the target bone part based on the three-dimensional image data of the target bone part; Overlaying the two-dimensional contour data of the target bone part onto the two-dimensional projection image of the target bone segment to generate a target image marked with the contour of the target bone part.

2. The method according to claim 1, wherein The determining the two-dimensional contour data of the target bone part based on the three-dimensional image data of the target bone part includes: Performing projection processing on the three-dimensional image data of the target bone part to obtain a two-dimensional projection image of the target bone part, where the projection direction of the two-dimensional projection image of the target bone part is consistent with the projection direction of the two-dimensional projection image of the target bone segment; Performing binarization processing on the two-dimensional projection image of the target bone part to obtain a binary image of the target bone part; Identifying the contours in the binary image, determining the target contour of the target bone part as the contour that meets the preset conditions, and obtaining the contour data of the target contour.

3. The method according to claim 2, wherein When the target bone segment is a vertebral segment and the projection direction is a frontal projection, the target bone part includes a spinous process and / or a pedicle; The identifying the contours in the binary image and determining the target contour of the target bone part as the contour that meets the preset conditions includes: Identifying all the contours in the binary image corresponding to the spinous process, and determining the contour with the largest area among all the contours as the target contour of the spinous process; And / or Identifying all the contours in the binary image corresponding to the pedicle, screening out the contours of the left pedicle and the right pedicle according to the area size, performing pixel brightness search within a preset radius based on the centroids of the contours of the left pedicle and the right pedicle, and determining the target contour of the pedicle according to the set of pixel points that reach the preset brightness searched out.

4. The method according to claim 2, wherein When the target bone segment is a sacral segment and the projection direction is a frontal projection, the target bone part includes the upper edge of the vertebral body and / or the sacral spinous process; The identifying the contours in the binary image and determining the target contour of the target bone part as the contour that meets the preset conditions includes: Identifying all the contours in the binary image corresponding to the vertebral body, and screening out the contour with the largest area among all the contours; Obtaining the two-dimensional projection image of the region corresponding to the contour with the largest area; Determining the target contour of the vertebral body according to the set of pixel points that reach the preset brightness in the preset direction in the two-dimensional projection image of the region.

5. The method according to claim 2, wherein When the target bone segment is a sacral segment and the projection direction is a lateral projection, the target bone part includes the vertebral body; The identifying the contours in the binary image and determining the target contour of the target bone part as the contour that meets the preset conditions includes: Identify all the contours in the binary image corresponding to the upper edge of the vertebral body, select the contour with the largest area among all the contours, segment the contour with the largest area, and determine the segmented contour as the target contour of the upper edge of the vertebral body; Perform skeletonization on the binary image corresponding to the sacral spinous process to obtain the target contour of the sacral spinous process.

6. The method according to claim 2, wherein When the target bone segment is the sacrum and the projection direction is the lateral projection, the target bone part includes the lateral side of the sacrum and / or the arc surface of the sacrum; The identifying the contours in the binary image and determining the contours that meet the preset conditions as the target contours of the target bone part includes: Identify all the contours in the binary image corresponding to the lateral side of the sacrum, select the contour with the largest area among all the contours, calculate the convex hull of the contour with the largest area, and based on the longest side of the convex hull, determine the set of pixel points with a preset brightness inside the binary image as the target contour of the lateral side of the sacrum; Identify the contours in the binary image corresponding to the arc surface of the sacrum, select the contour with the largest area among all the contours, perform pixel brightness search within a preset radius based on the centroid of the contour with the largest area, and determine the set of pixel points with the searched preset brightness as the target contour of the arc surface of the sacrum.

7. The method according to any one of claims 1 to 6, characterized in that The performing segmentation processing on the three-dimensional image data of the target bone segment to determine the three-dimensional image data of the target bone part includes: Determine the region range corresponding to the target bone part in the three-dimensional image data of the target bone segment; Generate the binary feature data of the target bone part according to the region range corresponding to the target bone part; Use the binary feature data of the target bone part to segment the three-dimensional image data of the target bone part from the three-dimensional image data of the target bone segment.

8. The method according to claim 7, wherein When the target bone segment is the vertebral segment, the target bone part includes the pedicle; The determining the region range corresponding to the target bone part in the three-dimensional image data of the target bone segment includes: Calculate the center line of the pedicle channel based on the characteristics of the vertebral segment; Based on the center line of the pedicle channel, respectively locate the region range corresponding to the left pedicle and the region range corresponding to the right pedicle in the three-dimensional image data of the vertebral segment; The generating the binary feature data of the target bone part according to the region range corresponding to the target bone part includes: Generate the first binary feature data according to the region range corresponding to the left pedicle; Generate the second binary feature data according to the region range corresponding to the right pedicle; Based on the Boolean algorithm, merge the first binary feature data and the second binary feature data to obtain the binary feature data of the pedicle.

9. The method according to claim 8, wherein When the target bone segment is the vertebral segment, the target bone part includes the spinous process; The determining the region range corresponding to the target bone part in the three-dimensional image data of the target bone segment includes: Construct multiple planes based on the center line of the pedicle channel, and locate the region range corresponding to the spinous process in the three-dimensional image data of the target bone segment according to the multiple planes and the preset displacement parameters.

10. The method according to claim 9, wherein When the target bone segment is a vertebral segment, the target bone part includes the vertebral body; Determining the region range corresponding to the target bone part in the three-dimensional image data of the target bone segment includes: Based on the center line of the pedicle channel, the region range corresponding to the spinous process, and the region range corresponding to the pedicle, locating the region range corresponding to the vertebral body in the three-dimensional image data of the vertebral segment.

11. The method according to claim 7, wherein When the target bone segment is the sacrum, the target bone part includes the left arc surface and the right arc surface; Determining the region range corresponding to the target bone part in the three-dimensional image data of the target bone segment includes: Calculating the sacral coordinate system based on the characteristics of the sacrum; Creating a first cube based on the sacral coordinate system and the position of the left sacral foramen, and determining the first cube as the region range corresponding to the left arc surface; Creating a second cube based on the sacral coordinate system and the position of the right sacral foramen, and determining the second cube as the region range corresponding to the right arc surface.

12. The method according to claim 11, characterized in that, When the target bone segment is the sacrum, the target bone part includes the sacral spinous process and the sacral vertebral body; Determining the region range corresponding to the target bone part in the three-dimensional image data of the target bone segment includes: Constructing a plurality of planes based on the sacral coordinate system, and based on the plurality of planes, locating the region ranges corresponding to the sacral spinous process and the sacral vertebral body in the three-dimensional image data of the sacrum.

13. An image processing apparatus, characterized in that, Including: An acquisition module for acquiring the three-dimensional image data and the two-dimensional projection image of the target bone segment, where the two-dimensional projection image of the target bone segment is obtained by performing projection processing on the three-dimensional image data of the target bone segment; An analysis module for determining the target bone part in the two-dimensional projection image of the target bone segment according to the projection direction of the two-dimensional projection image of the target bone segment; A segmentation module for segmenting the three-dimensional image data of the target bone segment to obtain the three-dimensional image data of the target bone part; A processing module for determining the contour data of the target bone part based on the three-dimensional image data of the target bone part; A synthesis module for superimposing the contour data of the target bone part onto the two-dimensional projection image of the target bone segment to generate a target image marked with the contour of the target bone part.

14. A computer device, characterized in that, The computer device includes: a processor and a memory storing computer program instructions; the processor reads and executes the computer program instructions to implement the image processing method according to any one of claims 1 to 12.

15. A computer-readable storage medium, characterized in that, A program or instructions are stored on the computer-readable storage medium, and when the program or instructions are executed by a processor, the image processing method according to any one of claims 1 to 12 is implemented.

16. A computer program product, characterized in that, When the instructions in the computer program product are executed by the processor of the computer device, the computer device is caused to execute the image processing method according to any one of claims 1 to 12.

Images