A method and apparatus for three-dimensional broken bone segmentation based on scribbles

Abstract

The application discloses a three-dimensional broken bone segmentation method and device based on picture painting, which comprises the following steps: firstly, reading existing three-dimensional CT bone segmentation data; secondly, mapping three-dimensional pixel centers in space to a screen; recording the coordinates of a mouse on the screen in each frame during the painting segmentation; calculating whether the three-dimensional pixel center is located in the painting range after the painting is completed, and recording the modified coordinates; after the painting is completed, the modified pixels are transmitted to a back end, the back end re-performs 3D modeling according to the modified segmentation; and finally, the model after the painting segmentation is displayed. The application can be applied to more accurate three-dimensional broken bone segmentation, can more intuitively display the segmentation process and the effect after the segmentation is completed, helps doctors to better complete diagnosis and understand the illness of patients, and provides a possibility for preoperative simulation planning and treatment.

Description

Technical Field

[0001] This invention belongs to the field of medical image processing and relates to a method and device for three-dimensional bone segmentation of medical CT bone images, which is particularly used to help doctors with preoperative planning, surgical simulation and treatment. Background Technology

[0002] Medical image processing is a rapidly developing field in recent years, with bone segmentation being a crucial aspect that plays a significant role in assisting medical diagnosis. As orthopedic surgical techniques advance, solutions utilizing medical image processing technologies to aid in surgical simulations are also gradually emerging. More and more doctors are using medical image processing techniques to assist in diagnosis and treatment, improving the safety and effectiveness of surgeries.

[0003] With the rapid development of computer technology, medical image processing technology has been widely applied in many fields, providing various technical solutions for bone segmentation. However, existing bone segmentation techniques still have certain limitations and cannot be applied to all situations. Currently, the mainstream bone segmentation methods are mainly divided into two types: traditional methods and deep learning methods. Traditional methods include threshold segmentation, region growing, and boundary tracking, but these methods all have certain limitations. For example, threshold segmentation is susceptible to noise and external interference, and region growing methods require manual specification of growth points. Deep learning methods, on the other hand, suffer from problems such as difficult data annotation, low training efficiency, and high computational costs. In addition, existing methods are difficult to meet the needs of bone segmentation when dealing with some special cases or when doctors require special treatment plans.

[0004] Therefore, at present, processing bone segmentation still requires some manual methods to improve the accuracy of bone segmentation and quickly and intuitively reflect the segmentation results. This invention starts with image smearing, manually applying different colored blocks to the screen surface to reflect the segmented bone entity, thus meeting different segmentation needs in various scenarios. In terms of algorithm, the mapping of smeared areas to 3D spatial entities is optimized by mapping the CT pixel coordinates of the bone entity to the screen, thereby avoiding overly complex spatial geometric calculations. Simultaneously, the calculation of smearing judgment is optimized by grouping CT pixels and using a divide-and-conquer method to accelerate the algorithm's running speed, allowing for faster and more intuitive modification of the segmentation model. Therefore, this invention has the characteristics of high accuracy, wide applicability, and fast processing speed, possessing high application value and promising prospects. It can provide a more efficient and accurate processing method for bone segmentation in multiple medical fields such as orthopedics and dentistry. This will better improve the diagnostic accuracy and treatment success rate of doctors, promote the development of the medical image processing field, and has significant innovative and practical value. Summary of the Invention

[0005] This invention aims to address the shortcomings of existing bone segmentation technologies, such as low accuracy, high computational cost, and limited applicability, by proposing a three-dimensional bone segmentation method and device based on image smearing.

[0006] The first aspect of this invention provides a three-dimensional bone segmentation method based on image smearing, comprising the following steps:

[0007] S1: Input a 3D CT skeleton image segmentation 3D array.

[0008] S2: Display the skeletal model in 3D space and show it on the screen.

[0009] S3: Record the mouse coordinates during the painting process, and draw a circle with radius r and semi-transparent color on the screen with the mouse coordinates as the center, representing the painted area.

[0010] S4: After the smearing is complete, modify the corresponding segmented 3D array.

[0011] S41: First, merge all pixels of the CT skeleton image into pixel groups of adjacent size s*s*s, where s represents the side length of the pixel group.

[0012] S42: For each pixel group, calculate its spatial center coordinates, project the coordinates onto the current screen, and find the corresponding position coordinates on the screen coordinate system.

[0013] S43: If the mouse coordinates are recorded in S3, and there exists a coordinate such that the distance from the center coordinate of the pixel group on the screen coordinate system is less than r+c, it is necessary to determine whether each pixel in the pixel group has been painted over.

[0014] S44: If the mouse coordinates recorded in S3 do not exist such that the distance from the center coordinate of the pixel group on the screen coordinate system is less than r+c, it means that no pixels in the pixel group have been painted over.

[0015] S431: For each pixel in the pixel group, calculate its spatial center coordinates, project the coordinates onto the current screen, and find the corresponding position coordinates on the screen coordinate system.

[0016] S432: If the coordinates of the mouse recorded in S3 exist such that the distance from the center coordinate of the pixel in the screen coordinate system is less than r, then modify the corresponding segmented three-dimensional array to the number of the color to be painted, and record the corresponding three-dimensional array index.

[0017] S433: If the coordinates of the mouse recorded in S3 do not exist such that the distance from the center coordinate of the pixel in the screen coordinate system is less than r, then no modification will be made.

[0018] S5: Send the 3D array index and the color number recorded in S432 to the backend.

[0019] S6: The backend modifies the 3D CT skeleton image by segmenting the 3D array based on the received 3D array index and the number of the painted color.

[0020] S7: After the front-end completes all the painting work, it sends a command to the back-end to remodel the skeletal model.

[0021] S8: Based on the modified segmented 3D array and the original segmented 3D array, determine the modified bone number and remodel this part of the bone.

[0022] S9: After the remodeling is completed, the smeared bone segmentation model is displayed in 3D space and shown on the screen.

[0023] Preferably, the value of s in step S41 is 4.

[0024] Preferably, in steps S43 and S44, if the CT pixel size is set to x*y*z, the c value is...

[0025] A second aspect of the present invention provides a three-dimensional bone segmentation device based on image smearing, comprising a memory and one or more processors, wherein the memory stores executable code, and the one or more processors execute the executable code to implement a three-dimensional bone segmentation method based on image smearing according to the present invention.

[0026] A third aspect of the invention relates to a computer-readable storage medium having a program stored thereon that, when executed by a processor, implements a three-dimensional bone segmentation method based on image smearing according to the invention.

[0027] This invention combines image smearing and 3D spatial mapping methods, applied to 3D CT images that have already undergone 3D modeling. By determining whether pixels in the 3D CT segmentation lie within the spatial mapping of the smeared area, the segmentation model of the fractured bone is modified. Finally, the segmented fractured bone is remodeled. First, existing 3D CT bone segmentation data is read in. Second, the centers of the 3D pixels in space are mapped onto the screen. During the smearing segmentation process, the mouse coordinates on the screen are recorded for each frame. After smearing, it is calculated whether the center of the 3D pixel lies within the smeared area, and the modified coordinates are recorded. After smearing, the modified pixels are passed to the backend, which remodels the 3D model based on the modified segmentation. Finally, the model after smearing and segmentation is displayed. This invention can be applied to more precise 3D fracture segmentation, providing a more intuitive display of the segmentation process and the final segmentation effect, helping doctors better diagnose and understand the patient's condition, and providing possibilities for preoperative simulation planning and treatment.

[0028] In summary, this invention creates a three-dimensional bone segmentation method based on screen smearing, which has the following beneficial effects: (1) During the smearing process, the bone model is segmented by mapping spatial CT pixel coordinates to the screen, thereby avoiding overly complex three-dimensional spatial calculations. (2) The calculation of smearing judgment is optimized by grouping CT pixels and using a divide-and-conquer method to speed up the algorithm's operation, allowing for faster and more intuitive modification of the segmented model. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 This is a flowchart of the method of the present invention.

[0031] Figure 2 yes Figure 1 The flowchart in the text describes the segmentation of a three-dimensional array for modifying the covered area. Specific implementation methods

[0032] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0033] Example 1

[0034] Reference Figure 1 and Figure 2 This example relates to a three-dimensional bone segmentation method based on image smearing.

[0035] Taking the process of smearing and segmenting a CT bone as an example, the method includes the following specific steps:

[0036] S1: Input a 3D CT skeleton image segmentation 3D array.

[0037] S2: Display the skeletal model in 3D space and show it on the screen.

[0038] S3: Record the mouse coordinates during the painting process, and draw a circle with radius r and semi-transparent color on the screen with the mouse coordinates as the center, representing the painted area.

[0039] S4: After the smearing is complete, modify the corresponding segmented 3D array.

[0040] S41: First, merge all pixels of the CT skeleton image into pixel groups of adjacent size s*s*s, where s represents the side length of the pixel group.

[0041] S42: For each pixel group, calculate its spatial center coordinates, project the coordinates onto the current screen, and find the corresponding position coordinates on the screen coordinate system.

[0042] S43: If the mouse coordinates are recorded in S3, and there exists a coordinate such that the distance from the center coordinate of the pixel group on the screen coordinate system is less than r+c, it is necessary to determine whether each pixel in the pixel group has been painted over.

[0043] S44: If the mouse coordinates recorded in S3 do not exist such that the distance from the center coordinate of the pixel group on the screen coordinate system is less than r+c, it means that no pixels in the pixel group have been painted over.

[0044] S431: For each pixel in the pixel group, calculate its spatial center coordinates, project the coordinates onto the current screen, and find the corresponding position coordinates on the screen coordinate system.

[0045] S432: If the coordinates of the mouse recorded in S3 exist such that the distance from the center coordinate of the pixel in the screen coordinate system is less than r, then modify the corresponding segmented three-dimensional array to the number of the color to be painted, and record the corresponding three-dimensional array index.

[0046] S433: If the coordinates of the mouse recorded in S3 do not exist such that the distance from the center coordinate of the pixel in the screen coordinate system is less than r, then no modification will be made.

[0047] S5: Send the 3D array index and the color number recorded in S432 to the backend.

[0048] S6: The backend modifies the 3D CT skeleton image by segmenting the 3D array based on the received 3D array index and the number of the painted color.

[0049] S7: After the front-end completes all the painting work, it sends a command to the back-end to remodel the skeletal model.

[0050] S8: Based on the modified segmented 3D array and the original segmented 3D array, determine the modified bone number and remodel this part of the bone.

[0051] S9: After the remodeling is completed, the smeared bone segmentation model is displayed in 3D space and shown on the screen.

[0052] Example 2

[0053] This embodiment provides a three-dimensional bone segmentation device based on image smearing, including a memory and one or more processors. The memory stores executable code, and when the one or more processors execute the executable code, they are used to implement a three-dimensional bone segmentation method based on image smearing according to Embodiment 1.

[0054] Example 3

[0055] This embodiment relates to a computer-readable storage medium storing a program that, when executed by a processor, implements a three-dimensional bone segmentation method based on image smearing as described in Embodiment 1.

[0056] The embodiments described in this specification are merely examples of implementations of the inventive concept. The scope of protection of this invention should not be considered as limited to the specific forms stated in the embodiments. The scope of protection of this invention also extends to equivalent technical means that can be conceived by those skilled in the art based on the inventive concept.

Claims

1. A three-dimensional bone segmentation method based on image smearing, characterized in that, Includes the following steps: S1: Input a 3D CT skeleton image segmentation 3D array; S2: Display the skeletal model in 3D space and show it on the screen; S3: Record the mouse coordinates during the painting process, and draw a semi-transparent circle with radius r on the screen centered at the mouse coordinates to represent the painted area; S4: After the smearing is completed, modify the corresponding segmentation 3D array; S41: First, divide all pixels of the CT skeleton image into pixel groups of size s*s*s, where s represents the side length of the pixel group; S42: For each pixel group, calculate its spatial center coordinates, project the coordinates onto the current screen, and find the corresponding position coordinates on the screen coordinate system; S43: If the mouse coordinates recorded in step S3 exist such that the distance from the center coordinate of the pixel group on the screen coordinate system is less than r+c, it is necessary to determine whether each pixel in the pixel group has been painted. S44: If the mouse coordinates recorded in step S3 do not exist such that the distance between the coordinates and the center coordinates of the pixel group on the screen coordinate system is less than r+c, it means that no pixels in the pixel group have been painted. S431: For each pixel in the pixel group, calculate its spatial center coordinates, project the coordinates onto the current screen, and find the corresponding position coordinates on the screen coordinate system. S432: If the coordinates of the mouse recorded in step S3 exist such that the distance between the coordinates and the center coordinates of the pixel on the screen coordinate system is less than r, then modify the corresponding segmented three-dimensional array to the number of the color to be painted, and record the corresponding three-dimensional array index. S433: If the mouse coordinates recorded in step S3 do not exist such that the distance between the coordinates and the center coordinates of the pixel in the screen coordinate system is less than r, then no modification will be made; S5: Send the three-dimensional array index and the number of the applied color recorded in step S432 to the backend; S6: The backend modifies the 3D CT skeleton image by segmenting the 3D array based on the received 3D array index and the number of the painted color; S7: After the front end completes all the painting work, it sends a command to the back end to remodel the skeletal model; S8: Based on the modified segmented 3D array and the original segmented 3D array, determine the modified bone number and remodel this part of the bone. S9: After the remodeling is completed, the smeared bone segmentation model is displayed in 3D space and shown on the screen.

2. The three-dimensional bone segmentation method based on image smearing according to claim 1, characterized in that: The value of s in step S41 is 4.

3. The three-dimensional bone segmentation method based on image smearing according to claim 1, characterized in that: In steps S43 and S44, if the CT pixel size is set to x*y*z, and the c value is...

4. A three-dimensional bone segmentation device based on image smearing, characterized in that, The device includes a memory and one or more processors, wherein the memory stores executable code, and the one or more processors execute the executable code to implement a three-dimensional bone segmentation method based on image smearing as described in any one of claims 1-3.

5. A computer-readable storage medium, characterized in that, It stores a program that, when executed by a processor, implements a three-dimensional bone segmentation method based on image smearing as described in any one of claims 1-3.

Images