A knee joint femur-tibia segmentation method and system based on region growing

Abstract

The present application relates to the technical field of medical image processing, and provides a knee femur and tibia segmentation method based on region growing, comprising the following steps: S1, acquiring knee medical images; S2, selecting a growth seed point; S3, coarsely segmenting the knee medical images to obtain knee stem data; S4, finely segmenting the knee medical images to obtain knee connection data; S5, fusing the knee stem data and the knee connection data to obtain knee femur and tibia image data with complete contours; S6, coarsely filling the knee femur and tibia image data to obtain knee femur and tibia image data after coarse filling; and S7, iteratively filling holes in the knee femur and tibia image data by using a three-dimensional sliding window, so as to obtain image data that is consistent with the surrounding data in shape, position and completeness and smoothness. The present application solves the problems of slow segmentation speed, unstable segmentation effect, incomplete segmentation and adhesion in preoperative planning of knee replacement surgery.

Description

Technical Field

[0001] This invention relates to the field of medical image processing technology, and in particular to a method and system for segmenting the femur and tibia of the knee joint based on region growth. Background Technology

[0002] The femur and tibia at the knee joint are the two most important bones connecting the entire knee joint. The separation of the femur and tibia is an indispensable step in knee replacement surgery. In the preoperative planning of knee replacement surgery, it is necessary to accurately and quickly separate the femur and tibia.

[0003] Most existing technologies use deep learning methods to segment the femur and tibia. However, this method has problems such as slow running speed, high computer configuration requirements, and instability in segmentation of some complex cases. Traditional segmentation algorithms, on the other hand, have problems such as incomplete segmentation and adhesion. Summary of the Invention

[0004] To address the aforementioned problems, the present invention aims to provide a method and system for femoral-tibial segmentation of the knee joint based on regional growth, in order to solve the problems of slow segmentation speed, unstable segmentation effect, incomplete segmentation, and adhesion in the preoperative planning of knee replacement surgery.

[0005] The above-mentioned objective of this invention is achieved through the following technical solutions:

[0006] A method for segmenting the femur and tibia of the knee joint based on region growth includes the following steps:

[0007] S1: Acquire medical images of the knee joint to be segmented;

[0008] S2: Manually select the growth seed point of the femur or tibia on the medical image of the knee joint;

[0009] S3: Perform coarse segmentation on the knee joint medical image to obtain knee joint bone data after coarse segmentation;

[0010] S4: Perform fine segmentation on the knee joint medical image, optimize the growth conditions in the region growth algorithm, supplement the missing knee joint data in the knee joint bone data after coarse segmentation, and obtain the knee joint connection data at the femoral and tibial junction after fine segmentation.

[0011] S5: Perform data fusion on the knee joint bone data after coarse segmentation and the knee joint connection data after fine segmentation to obtain knee joint femoral tibia image data with complete contours;

[0012] S6: Perform coarse filling on the fused femoral-tibial image data of the knee joint to fill in the data points located inside the bone, and obtain the femoral-tibial image data of the knee joint after coarse filling.

[0013] S7: The knee joint femoral tibia image data after coarse filling is filled with holes iteratively using a three-dimensional sliding window to obtain knee joint femoral tibia image data that is consistent with the shape and position of the surrounding data and is complete and smooth.

[0014] Further, in step S3, the knee joint medical image is coarsely segmented to obtain the knee joint bone shaft data after coarse segmentation, specifically as follows:

[0015] A first HU value range is set, and data points in the knee joint medical imaging data that fall within the first HU value range are added to the growth queue, and the data points are marked as three-dimensional model points;

[0016] A second HU value range is set, and the data points in the knee joint medical imaging data that fall within the second HU value range are marked as the three-dimensional model points, but are not added to the growth queue;

[0017] Wherein, the first HU value range and the second HU value range are the HU value ranges representing bone in the knee joint medical imaging data, and the HU value in the first HU value range is greater than the HU value in the second HU value range.

[0018] Further, in step S4, the knee joint medical image is subjected to fine segmentation, the growth conditions in the region growing algorithm are optimized, and the missing knee joint data in the knee joint bone data after coarse segmentation is supplemented to obtain the knee joint connection data at the femoral-tibial junction after fine segmentation. Specifically:

[0019] Starting from the seed point, calculate the average HU value of the 3*3*3 spatial domain for each point to be grown in the growth queue;

[0020] If the HU value of the growth point to be grown is higher than the average HU value, and it is determined that the growth point to be grown is not located at the junction of the femur and tibia, the growth point to be grown is added to the growth queue and marked as the three-dimensional model point.

[0021] The growth queue is iterated continuously to obtain the knee joint connection data at the junction of the femur and tibia after the fine segmentation.

[0022] Specifically, determining that the growth point is not located at the junction of the femur and tibia involves:

[0023] Based on the relationship between layers in the knee joint medical imaging, when the HU value of the growth point is higher than the HU value of the upper layer or the lower layer, the growth point is not located at the junction of the femur and tibia.

[0024] Further, in step S5, the knee joint shaft data obtained through the coarse segmentation and the knee joint connection data obtained through the fine segmentation are fused to obtain knee joint femoral-tibial image data with complete contours, specifically:

[0025] The three-dimensional model points in the knee joint bone data after coarse segmentation and the three-dimensional marker points marked in the knee joint connection data after fine segmentation are obtained respectively.

[0026] The three-dimensional model points in the knee joint bone data and the three-dimensional marker points marked in the knee joint connection data are merged and fused. The image data composed of all the three-dimensional marker points after merging and fusion is the knee joint femoral tibia image data with complete outline.

[0027] Further, in step S6, the fused knee joint femoral-tibial image data is coarsely filled to fill the data points located inside the bone, thereby obtaining the knee joint femoral-tibial image data after coarse filling, specifically as follows:

[0028] Determine whether the data points in the knee joint femoral and tibial imaging data have marked three-dimensional model points in all three directions (X, Y, Z). If the marked three-dimensional model points exist in all three directions (X, Y, Z), then the data points are considered to be located inside the bone and are marked as three-dimensional model points.

[0029] By iterating through the data points in the knee joint femoral-tibial image data, the knee joint femoral-tibial image data after coarse filling is obtained.

[0030] Further, in step S7, the knee joint femoral-tibial image data after coarse filling is iteratively filled with holes using the three-dimensional sliding window to obtain knee joint femoral-tibial image data that is consistent with the surrounding data in shape, position, and is complete and smooth. Specifically:

[0031] Construct a three-dimensional sliding window and place the three-dimensional sliding window within the knee joint femoral tibia image data after coarse filling;

[0032] The three-dimensional sliding window is slid sequentially. When the data point corresponding to the three-dimensional center point in the three-dimensional sliding window is not marked as the three-dimensional model point, the number of overlapping directions of the three-dimensional marker points between the three-dimensional sliding window and the knee joint femoral tibia image data after coarse filling is determined. If the number of overlapping directions is greater than 4, the data point is marked as the three-dimensional model point and added to the growth queue.

[0033] Through continuous iteration, the knee joint femoral tibia imaging data is obtained that is consistent with the surrounding data in terms of shape, position, and completeness and smoothness.

[0034] A region-growing-based knee joint femoral-tibial segmentation system for performing the region-growing-based knee joint femoral-tibial segmentation method as described above, comprising:

[0035] The image acquisition module is used to acquire medical images of the knee joint to be segmented.

[0036] The seed point selection module is used to manually select growth seed points of the femur or tibia on the medical image of the knee joint.

[0037] The coarse segmentation module is used to perform coarse segmentation on the knee joint medical image to obtain knee joint bone data after coarse segmentation.

[0038] The fine segmentation module is used to perform fine segmentation on the knee joint medical image, optimize the growth conditions in the region growth algorithm, supplement the missing knee joint data in the knee joint bone data after the coarse segmentation, and obtain the knee joint connection data at the femoral and tibial junction after the fine segmentation.

[0039] The data fusion module is used to fuse the knee joint bone data after coarse segmentation and the knee joint connection data after fine segmentation to obtain knee joint femoral and tibia image data with complete contours.

[0040] The coarse fill module is used to coarsely fill the fused knee joint femoral tibia image data, filling in the data points located inside the bone, to obtain the knee joint femoral tibia image data after coarse filling.

[0041] The fine filling module is used to iteratively fill the holes in the knee joint femoral tibia image data after the coarse filling using a three-dimensional sliding window, so as to obtain the knee joint femoral tibia image data that is consistent with the shape and position of the surrounding data and is complete and smooth.

[0042] A computer device includes a memory and one or more processors, the memory storing computer code that, when executed by the one or more processors, causes the one or more processors to perform the method described above.

[0043] A computer-readable storage medium storing computer code that, when executed, performs the method described above.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] This invention provides a method for femoral-tibial segmentation of the knee joint based on region growing, comprising: S1: acquiring a medical image of the knee joint to be segmented; S2: manually selecting growth seed points for the femur or tibia on the medical image of the knee joint; S3: performing coarse segmentation on the medical image of the knee joint to obtain knee joint shaft data after coarse segmentation; S4: performing fine segmentation on the medical image of the knee joint, optimizing the growth conditions in the region growing algorithm, supplementing the missing knee joint data in the knee joint shaft data after coarse segmentation, and obtaining knee joint connection data at the junction of the femur and tibia after fine segmentation. S5: The knee joint bone data obtained through coarse segmentation and the knee joint connection data obtained through fine segmentation are fused to obtain knee joint femoral-tibial image data with complete contours; S6: The fused knee joint femoral-tibial image data is coarsely filled to fill data points located inside the bone, obtaining knee joint femoral-tibial image data after coarse filling; S7: The knee joint femoral-tibial image data after coarse filling is iteratively filled with holes using a three-dimensional sliding window to obtain knee joint femoral-tibial image data with consistent shape, position, and complete smoothness with the surrounding data. The above technical solution, by using region growing method and computer image processing technology to process medical image data, solves the problems of segmentation loss and adhesion in current traditional methods and the slow segmentation speed of deep learning methods, achieving stable and fast knee joint segmentation, providing a better foundation for knee replacement surgery. Attached Figure Description

[0046] Figure 1 A schematic diagram of the overall process of a knee joint femoral-tibial segmentation method based on region growth provided by the present invention;

[0047] Figure 2 A schematic diagram of the coarse segmentation method in a region-growing-based knee joint femoral-tibial segmentation method provided by the present invention;

[0048] Figure 3The segmentation effect diagram of the coarse segmentation method in the knee joint femoral-tibial segmentation method provided by the present invention;

[0049] Figure 4 A schematic diagram of the fine segmentation method in a region-growing-based knee joint femoral-tibial segmentation method provided by the present invention;

[0050] Figure 5 The segmentation effect diagram of the fine segmentation method in the knee joint femoral-tibial segmentation method provided by the present invention;

[0051] Figure 6 The image shows the fusion effect of coarse and fine segmentation data in a region-growing-based femoral-tibial segmentation method for the knee joint provided by this invention.

[0052] Figure 7 A schematic diagram of the coarse filling process of the femur and tibia in a region-growing-based femoral-tibial segmentation method for the knee joint provided by the present invention;

[0053] Figure 8 A rough filling effect diagram of a region-growing-based method for segmenting the femur and tibia of the knee joint provided by this invention;

[0054] Figure 9 A schematic diagram of the process of fine filling of the femoral and tibia in a region-growing-based femoral-tibial segmentation method for the knee joint provided by the present invention;

[0055] Figure 10 A fine-filling effect diagram in a knee joint femoral-tibial segmentation method based on region growth provided by the present invention;

[0056] Figure 11 This invention provides an overall structural diagram of a knee joint femoral-tibial segmentation system based on region growth. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0058] Those skilled in the art will understand that, unless specifically stated otherwise, the singular forms “a,” “an,” “the,” and “the” used herein may also include the plural forms. It should be further understood that the term “comprising” as used in this specification means the presence of the stated features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0059] First Embodiment

[0060] like Figure 1 As shown, this embodiment provides a method for segmenting the femoral and tibia of the knee joint based on region growth, including the following steps:

[0061] S1: Obtain medical images of the knee joint to be segmented.

[0062] Specifically, the knee joint medical images involved in this invention can be acquired using any type of professional medical imaging equipment used in hospitals. For example, medical images acquired using X-ray projection equipment, CT projection equipment, etc. This invention does not impose any restrictions on the method of acquiring knee joint medical images.

[0063] S2: Manually select the growth seed point of the femur or tibia on the medical image of the knee joint.

[0064] Specifically, because each patient's knee joint medical imaging data is different, doctors need to select growth seed points on the femur and tibia respectively, based on each patient's characteristics, for subsequent segmentation and filling processes.

[0065] S3: Perform coarse segmentation on the knee joint medical images to obtain knee joint bone shaft data after coarse segmentation, specifically:

[0066] A first HU value range is set, and data points in the knee joint medical imaging data that fall within the first HU value range are added to a growth queue and marked as 3D model points. A second HU value range is set, and data points in the knee joint medical imaging data that fall within the second HU value range are marked as 3D model points, but are not added to the growth queue. The first HU value range and the second HU value range are the HU value ranges representing bone in the knee joint medical imaging data, and the HU values ​​in the first HU value range are greater than the HU values ​​in the second HU value range.

[0067] In this embodiment, the first HU value range is set to a HU value higher than 700, and the second HU value range is set to a HU value between 400 and 700. HU value is a unit of measurement used in medical imaging to determine the density of human tissues or organs. The HU value of bone differs significantly from other tissues, generally exceeding 400. Based on this characteristic, the coarse growth conditions in the region growth are divided into two parts: the first part, data points with a HU value higher than 700 (the first HU value range) are added to the growth queue first and marked as 3D model points; the second part, data points with a HU value between 400 and 700 (the second HU value range) are marked as 3D model points but not added to the growth queue to avoid overgrowth. This obtains knee joint bone shaft data with distinct characteristics. Figure 2 This is a schematic diagram of the coarse segmentation method in a region-growing-based method for segmenting the femur and tibia of the knee joint provided by the present invention. Figure 3 The image shows the segmentation effect of the coarse segmentation method in the knee joint femoral-tibial segmentation method provided in this application.

[0068] S4: Perform fine segmentation on the knee joint medical image, optimize the growth conditions in the region growing algorithm, supplement the missing knee joint data in the knee joint bone data after the coarse segmentation, and obtain the knee joint connection data at the femoral-tibial junction after the fine segmentation, specifically:

[0069] In order to optimize the adhesion problem in traditional segmentation algorithms, this application makes full use of the position information of each data point, and takes the growth seed point as the starting point to calculate the average HU value of the 3*3*3 spatial domain of each point to be grown in the growth queue.

[0070] Using the average HU (Heterogeneous Humility) value of the spatial neighborhood corresponding to each growth point as a reference standard, if the HU value of a data point is lower than the average HU value of the spatial neighborhood, it indicates that the data point may be located at the edge of the bone or at the junction of bones; if the HU value of the growth point is higher than the average HU value, it indicates that the data point is located on the bone, and proceed to the next step. This method can effectively avoid most adhesions by adaptively adjusting the reference standard of data points at different locations. However, due to the special complexity of patient knee joint medical imaging data, noise and other problems may exist. Therefore, in the next step, the relationship between layers in the knee joint medical imaging data is fully utilized. When the growth point is located at the junction of the femur and tibia, there will be a minimum point in the HU value distribution of this point in the vertical direction of the sagittal plane. That is, when the HU value of the growth point is higher than the HU value of the upper femur point or higher than the HU value of the lower tibia point, it is considered that the growth point will not be located at the minimum point of the femur-tibia junction. The growth point is then added to the growth queue and marked as a point in the three-dimensional model.

[0071] The growth queue is iterated continuously to obtain the knee joint connection data at the junction of the femur and tibia after the fine segmentation. Figure 4 This application provides a schematic diagram of the fine segmentation method in a region-growing-based method for segmenting the femur and tibia of the knee joint. Figure 5 The image shows the segmentation effect of the fine segmentation method in the knee joint femoral-tibial segmentation method provided in this application.

[0072] S5: The knee joint bone data obtained through coarse segmentation and the knee joint connection data obtained through fine segmentation are fused to obtain knee joint femoral-tibial image data with complete contours, specifically:

[0073] The three-dimensional model points in the knee joint bone data after coarse segmentation and the three-dimensional marker points in the knee joint connection data after fine segmentation are obtained respectively. The three-dimensional model points in the knee joint bone data and the three-dimensional marker points in the knee joint connection data are merged and fused. The image data composed of all the three-dimensional marker points after merging and fusion is the image data of the femoral and tibia of the knee joint with complete outline.

[0074] Figure 6 This image shows the fusion of coarse and fine segmentation data in a region-growing-based femoral-tibial segmentation method for the knee joint provided in this application. This method effectively avoids femoral-tibial adhesions while simultaneously providing a relatively complete segmentation of the overall contours of the femur and tibia.

[0075] S6: Perform coarse filling on the fused knee joint femoral-tibial image data to fill in the data points located inside the bone, and obtain the knee joint femoral-tibial image data after coarse filling.

[0076] Specifically, in knee joint segmentation, due to the presence of noise points and holes, coarse filling is first performed to obtain more complete and smooth 3D data. This involves introducing the positional information of the data points and determining whether the data points in the knee joint femoral-tibial image data have marked 3D model points in all three directions (X, Y, and Z). If marked 3D model points exist in all three directions, the data points are considered to be located inside the bone and are marked as 3D model points. Through continuous iteration of the data points in the knee joint femoral-tibial image data, the knee joint femoral-tibial image data after coarse filling is obtained.

[0077] Figure 7 This is a flowchart illustrating the coarse filling process of the femoral and tibia in a region-growing-based femoral-tibial segmentation method for the knee joint provided in this application. Figure 8The image shows the effect of coarse filling in a method for segmenting the femur and tibia of the knee joint based on region growth provided in this application.

[0078] S7: The knee joint femoral-tibial image data after coarse filling is iteratively filled with holes using a three-dimensional sliding window to obtain knee joint femoral-tibial image data that is consistent with the surrounding data in shape and position, and is complete and smooth. Specifically:

[0079] Construct a three-dimensional sliding window and place the three-dimensional knee joint femoral tibia image data after coarse filling;

[0080] The three-dimensional sliding window is slid sequentially. When the data point corresponding to the three-dimensional center point in the three-dimensional sliding window is not marked as the three-dimensional model point, the number of overlapping directions of the three-dimensional marker points between the three-dimensional sliding window and the knee joint femoral tibia image data after coarse filling is determined (the six directions of up, down, left, right, front, and back of the data point are determined sequentially). If the number of overlapping directions is greater than 4, the data point is marked as the three-dimensional model point and added to the growth queue. Through continuous iteration, the knee joint femoral tibia image data with the same shape, position and complete smoothness as the surrounding data is obtained.

[0081] Figure 9 This is a schematic diagram of the process of fine filling of the femoral and tibia in a region-growing-based femoral-tibial segmentation method for the knee joint provided in this application. Figure 10 A fine-filling effect diagram of a region-growing-based method for segmenting the femur and tibia of the knee joint provided in this application.

[0082] Second Embodiment

[0083] like Figure 11 As shown, this embodiment provides a region-growing-based knee joint femoral-tibial segmentation system for performing the region-growing-based knee joint femoral-tibial segmentation method as described in the first embodiment, comprising:

[0084] Image acquisition module 1 is used to acquire medical images of the knee joint to be segmented;

[0085] Seed point selection module 2 is used to manually select growth seed points of the femur or tibia on the knee joint medical image;

[0086] The coarse segmentation module 3 is used to coarsely segment the knee joint medical image to obtain knee joint bone data after coarse segmentation.

[0087] The fine segmentation module 4 is used to perform fine segmentation on the knee joint medical image, optimize the growth conditions in the region growth algorithm, supplement the missing knee joint data in the knee joint bone data after the coarse segmentation, and obtain the knee joint connection data at the femoral and tibial junction after the fine segmentation.

[0088] Data fusion module 5 is used to fuse the knee joint bone data after coarse segmentation and the knee joint connection data after fine segmentation to obtain knee joint femoral tibia image data with complete contours.

[0089] The coarse filling module 6 is used to coarsely fill the fused knee joint femoral tibia image data, filling in the data points located inside the bone, and obtaining the knee joint femoral tibia image data after coarse filling.

[0090] The fine filling module 7 is used to iteratively fill the holes in the knee joint femoral tibia image data after the coarse filling using a three-dimensional sliding window, so as to obtain the knee joint femoral tibia image data that is consistent with the shape and position of the surrounding data and is complete and smooth.

[0091] A computer-readable storage medium stores computer code that, when executed, performs the methods described above. Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. This program can be stored in a computer-readable storage medium, which may include: read-only memory (ROM), random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0092] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

[0093] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0094] It should be noted that the above embodiments can be freely combined as needed. The above description is only a preferred embodiment of the present invention. It should be pointed out that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for segmenting the femur and tibia of the knee joint based on region growth, characterized in that, Includes the following steps: S1: Acquire medical images of the knee joint to be segmented; S2: Manually select the growth seed point of the femur or tibia on the medical image of the knee joint; S3: Perform coarse segmentation on the knee joint medical image to obtain knee joint bone data after coarse segmentation; S4: Perform fine segmentation on the knee joint medical image, optimize the growth conditions in the region growing algorithm, supplement the missing knee joint data in the knee joint bone data after the coarse segmentation, and obtain the knee joint connection data at the femoral-tibial junction after the fine segmentation, specifically: Starting from the seed point, the average HU value of the 3*3*3 spatial neighborhood of each point to be grown in the growth queue is calculated. If the HU value of the point to be grown is higher than the average HU value, and it is determined that the point to be grown is not located at the junction of the femur and tibia, the point to be grown is added to the growth queue and marked as a three-dimensional marker point. The points to be grown in the growth queue are iterated continuously to obtain the knee joint connection data at the junction of the femur and tibia after the fine segmentation. S5: Perform data fusion on the knee joint bone data after coarse segmentation and the knee joint connection data after fine segmentation to obtain knee joint femoral tibia image data with complete contours; S6: Perform coarse filling on the fused femoral-tibial image data of the knee joint to fill in the data points located inside the bone, and obtain the femoral-tibial image data of the knee joint after coarse filling. S7: The knee joint femoral tibia image data after coarse filling is filled with holes iteratively using a three-dimensional sliding window to obtain knee joint femoral tibia image data that is consistent with the shape and position of the surrounding data and is complete and smooth.

2. The method for segmenting the femur and tibia of the knee joint based on region growth according to claim 1, characterized in that, In step S3, the knee joint medical image is coarsely segmented to obtain the knee joint bone shaft data after coarse segmentation, specifically as follows: A first HU value range is set, and data points in the knee joint medical imaging data that fall within the first HU value range are added to the growth queue, and the data points are marked as three-dimensional model points; A second HU value range is set, and the data points in the knee joint medical imaging data that fall within the second HU value range are marked as the three-dimensional model points, but are not added to the growth queue; Wherein, the first HU value range and the second HU value range are the HU value ranges representing bone in the knee joint medical imaging data, and the HU value in the first HU value range is greater than the HU value in the second HU value range.

3. The method for segmenting the femur and tibia of the knee joint based on region growth according to claim 2, characterized in that, Specifically, the determination that the growth point is not located at the junction of the femur and tibia is as follows: Based on the relationship between layers in the knee joint medical imaging, when the HU value of the growth point is higher than the HU value of the upper layer or the lower layer, the growth point is not located at the junction of the femur and tibia.

4. The method for segmenting the femur and tibia of the knee joint based on region growth according to claim 2, characterized in that, In step S5, the knee joint shaft data obtained through coarse segmentation and the knee joint connection data obtained through fine segmentation are fused to obtain knee joint femoral-tibial image data with complete contours, specifically: The three-dimensional model points in the knee joint bone data after coarse segmentation and the three-dimensional marker points marked in the knee joint connection data after fine segmentation are obtained respectively. The three-dimensional model points in the knee joint bone data and the three-dimensional marker points marked in the knee joint connection data are merged and fused. The image data composed of all the three-dimensional model points and three-dimensional marker points after merging and fusion is the knee joint femoral and tibia image data with complete outline.

5. The method for segmenting the femur and tibia of the knee joint based on region growth according to claim 4, characterized in that, In step S6, the fused knee joint femoral-tibial image data is coarsely filled to fill in the data points located inside the bone, thereby obtaining the knee joint femoral-tibial image data after coarse filling. Specifically: Determine whether the data points in the knee joint femoral and tibial imaging data have marked three-dimensional model points in all three directions (X, Y, Z). If the marked three-dimensional model points exist in all three directions (X, Y, Z), then the data points are considered to be located inside the bone and are marked as three-dimensional model points. By iterating through the data points in the knee joint femoral-tibial image data, the knee joint femoral-tibial image data after coarse filling is obtained.

6. The method for segmenting the femur and tibia of the knee joint based on region growth according to claim 5, characterized in that, In step S7, the three-dimensional sliding window is used to iteratively fill the holes in the knee joint femoral-tibial image data after the coarse filling, to obtain knee joint femoral-tibial image data that is consistent with the surrounding data in shape and position, and is complete and smooth. Specifically: Construct a three-dimensional sliding window and place the three-dimensional sliding window within the knee joint femoral tibia image data after coarse filling; The three-dimensional sliding window is slid sequentially. When the data point corresponding to the three-dimensional center point in the three-dimensional sliding window is not marked as the three-dimensional model point, the number of overlapping directions of the three-dimensional marker points between the three-dimensional sliding window and the knee joint femoral tibia image data after coarse filling is determined. If the number of overlapping directions is greater than 4, the data point is marked as the three-dimensional model point and added to the growth queue. Through continuous iteration, the knee joint femoral tibia imaging data is obtained that is consistent with the surrounding data in terms of shape, position, and completeness and smoothness.

7. A region-growing-based knee joint femoral-tibial segmentation system for performing the region-growing-based knee joint femoral-tibial segmentation method as described in any one of claims 1-6, characterized in that, include: The image acquisition module is used to acquire medical images of the knee joint to be segmented. The seed point selection module is used to manually select growth seed points of the femur or tibia on the medical image of the knee joint. The coarse segmentation module is used to perform coarse segmentation on the knee joint medical image to obtain knee joint bone data after coarse segmentation. The fine segmentation module is used to perform fine segmentation on the knee joint medical image, optimize the growth conditions in the region growth algorithm, supplement the missing knee joint data in the knee joint bone data after the coarse segmentation, and obtain the knee joint connection data at the femoral and tibial junction after the fine segmentation. The data fusion module is used to fuse the knee joint bone data after coarse segmentation and the knee joint connection data after fine segmentation to obtain knee joint femoral and tibia image data with complete contours. The coarse fill module is used to coarsely fill the fused knee joint femoral tibia image data, filling in the data points located inside the bone, to obtain the knee joint femoral tibia image data after coarse filling. The fine filling module is used to iteratively fill the holes in the knee joint femoral tibia image data after the coarse filling using a three-dimensional sliding window, so as to obtain the knee joint femoral tibia image data that is consistent with the shape and position of the surrounding data and is complete and smooth.

8. A computer device comprising a memory and one or more processors, the memory storing computer code that, when executed by the one or more processors, causes the one or more processors to perform the method as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing computer code, wherein when the computer code is executed, the method of any one of claims 1 to 6 is performed.

Images