Method, apparatus and computer equipment for generating acetabular registration points
By generating acetabular registration guide points in hip bone images, the problem of unreasonable registration point selection caused by difficulties in structural recognition or poor image quality in acetabular surgery is solved, achieving high-precision acetabular surgical registration and improving surgical efficiency and safety.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- YUANHUA ORTHOPAEDIC ROBOTICS (SHENZHEN) LTD
- Filing Date
- 2026-05-08
- Publication Date
- 2026-07-31
AI Technical Summary
In acetabular surgery, difficulties in identifying acetabular structures in preoperative images or poor image quality can lead to unreasonable or inaccurate selection of registration points, affecting the accuracy and reliability of the registration operation.
By generating acetabular registration guide points, computer equipment is used to identify the acetabulum in hip bone images and determine multiple internal and external registration guide points, guiding the surgeon to select reasonable and accurate registration points and improving registration accuracy.
It can reasonably and accurately select registration points under various imaging conditions, reduce the difficulty of the surgeon's work, ensure the reliability of registration results, and improve the efficiency and safety of acetabular surgery.
Smart Images
Figure CN122492773A_ABST
Abstract
Description
Technical Field
[0001] This application belongs to the field of computer-aided medical technology, and in particular relates to a method, apparatus and computer equipment for generating acetabular registration points. Background Technology
[0002] In some complex orthopedic surgeries, computer-aided navigation can effectively improve surgical precision and postoperative outcomes. The key to this technology lies in achieving high-precision registration between the patient's actual intraoperative anatomical structures and preoperative medical images.
[0003] Taking acetabular surgery as an example, high-precision registration requires the rational selection of registration points, which depends on the accurate identification and segmentation of the target anatomical structures. For instance, it is necessary to accurately identify the complete shape of the acetabulum and select registration points accordingly. However, in actual operation, situations often arise where the acetabular structure is difficult to identify or the image quality is poor in the preoperative images due to various reasons, which greatly affects the selection of registration points and poses a serious challenge to subsequent registration operations. Summary of the Invention
[0004] In view of this, embodiments of this application provide a method, apparatus, and computer device for generating acetabular registration points, which guides the selection of registration points by generating registration guide points, thereby improving the rationality and accuracy of registration point selection and improving the accuracy of subsequent registration operations.
[0005] The first aspect of this application provides a method for generating acetabular registration points, including: Acquire a patient's hip bone image and perform acetabular identification in the hip bone image to obtain an identification result, the identification result including the image quality of the image region where the acetabular is located in the hip bone image; Based on the identification result of the acetabulum, multiple registration guide points are determined, including multiple points located inside and / or outside the image region where the acetabulum is located; Based on the multiple registration guide points, multiple registration points are generated, and the multiple registration points are used for surface registration of the acetabulum.
[0006] Optionally, acquiring the patient's hip bone image and performing acetabular identification within the hip bone image to obtain the identification result includes: Acquire the patient's medical imaging data, and segment the hip bone image from the medical imaging data, which includes CT image data; A pre-trained acetabular recognition model is invoked to identify the hip bone image and determine the image region where the acetabulum is located in the hip bone image; wherein, the acetabular recognition model is trained based on multiple pre-labeled acetabular sample data, and the acetabular recognition model includes an image enhancement module, which is used to perform image enhancement processing on the image region where the acetabulum is located before the acetabular recognition model outputs the recognition result of the acetabulum.
[0007] Optionally, the image quality is characterized by image resolution, and the determination of multiple registration guide points based on the identification result of the acetabulum includes: When the image resolution of the image region where the acetabulum is located is less than the first resolution, multiple registration guide points are generated outside the image region. When the image resolution of the image region where the acetabulum is located is greater than or equal to the first resolution, multiple registration guide points are generated inside and outside the image region, respectively; wherein, the number of registration guide points located outside the image region is greater than the number of registration guide points located inside the image region.
[0008] Optionally, generating multiple registration guide points outside the image region includes: A first guiding region is determined outside the image region, and multiple candidate guiding points are generated within the first guiding region; The image region is fitted into a circle, and the circle is divided into four quadrants with its center. From a plurality of candidate guide points located in the intersection region of each of the quadrants and the first guide region, a plurality of registration guide points are uniformly determined; Multiple verification points are determined from a plurality of candidate guide points that have not been identified as the registration guide point. These verification points are used to verify the registration results after the acetabulum is registered.
[0009] Optionally, generating multiple candidate guide points in the first guide region includes: Acquire multiple reference hip images, each of which is pre-marked with multiple candidate guide points; The hip bone image is matched with multiple reference hip bone images to obtain the similarity between the hip bone image and each reference hip bone image; Based on multiple candidate guide points in the target reference hip image with a similarity greater than a preset value, multiple candidate guide points are determined in a first guide region outside the image region.
[0010] Optionally, the generation of multiple registration guide points both inside and outside the image region includes: Multiple anatomical landmarks are determined both inside and outside the image region; Multiple registration guide points are determined from the multiple anatomical landmarks.
[0011] Optionally, it also includes: If the acetabulum cannot be identified in the hip bone image, a manual operation prompt message is generated; In response to the image region marked in the hip bone image for the manual operation prompt information, a plurality of registration guide points are generated outside the marked image region.
[0012] Optionally, generating multiple registration points based on the multiple registration guide points includes: Multiple registration guide points are shown in the image used for acetabular registration; In response to multiple points selected in the image based on multiple registration guide points, as multiple registration points.
[0013] A second aspect of this application provides an acetabular registration point generation device, comprising: The hip bone image acquisition module is used to acquire images of the patient's hip bone. An acetabular recognition module is used to perform acetabular recognition in the hip bone image and obtain a recognition result, the recognition result including the image quality of the image region where the acetabular is located in the hip bone image; The registration guide point determination module is used to determine multiple registration guide points based on the recognition result of the acetabulum. The registration guide points include multiple points located inside and / or outside the image area where the acetabulum is located. The registration point generation module is used to generate multiple registration points based on multiple registration guide points, and the multiple registration points are used for surface registration of the acetabulum.
[0014] A third aspect of this application provides a computer device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the computer device performs the method described in the first aspect above.
[0015] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a computer, implements the method described in the first aspect above.
[0016] A fifth aspect of this application provides a computer program product, including a computer program that, when run, causes the method described in the first aspect to be executed.
[0017] Compared with the prior art, the embodiments of this application have the following beneficial effects: In this embodiment, a computer device acquires an image of the patient's hip bone and can identify the acetabulum within that image, obtaining corresponding identification results. These results may include information such as the image region containing the acetabulum and the image quality of that region. Based on the acetabular identification results, the computer device first determines multiple registration guide points. These guide points can be located inside and / or outside the image region containing the acetabulum. Thus, multiple registration points can be generated based on these guide points, allowing for planar registration of the acetabulum. This embodiment can, based on the obtained acetabular image, selectively choose appropriate processing methods according to different image qualities, generating registration guide points to guide the surgeon in rationally selecting registration points. For example, when the acetabulum image quality is poor or the acetabulum cannot be identified, the surgeon is guided to acquire stable and clear anatomical points outside the acetabulum, bypassing the problem area. This ensures the feasibility of registration in extreme cases and guarantees successful subsequent registration, avoiding the risk of being unable to continue surgery due to navigation failure. When the acetabulum image quality is good, by combining registration points from both the inner and outer parts of the acetabulum, and utilizing both local details and global frame information, rich spatial constraints are provided to the maximum extent, thereby ensuring high-precision registration results. Applying the methods, apparatus, and equipment provided in the embodiments of this application can improve the overall efficiency and safety of acetabular surgery. Attached Figure Description
[0018] To more clearly illustrate the technical solutions in the embodiments of this application, 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 this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0019] Figure 1 This is a schematic diagram of a method for generating acetabular registration points according to an embodiment of this application; Figure 2 This is a schematic diagram of another method for generating acetabular registration points provided in an embodiment of this application; Figure 3 This is a schematic diagram illustrating the generation of registration guide points provided in an embodiment of this application; Figure 4 This is a schematic diagram of an acetabular registration process provided in an embodiment of this application; Figure 5 This is a schematic diagram of an acetabular registration point generation device provided in an embodiment of this application; Figure 6 This is a schematic diagram of a computer device provided in an embodiment of this application. Detailed Implementation
[0020] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.
[0021] As described in the background section, in actual acetabular registration procedures, difficulties in identifying acetabular structures or poor image quality in preoperative images are frequently encountered, significantly impacting the selection of registration points and posing a severe challenge to subsequent registration operations. For example, pathological factors can cause changes or blurring of anatomical structures. In patients with severe osteoporosis, bone boundaries may be blurred on images, making them difficult to distinguish from surrounding soft tissues; complex acetabular fractures, especially comminuted fractures, can also disrupt the normal anatomical contour of the acetabulum, resulting in fracture fragment displacement and overlap; furthermore, acetabular dysplasia, severe morphological abnormalities in the acetabulum itself, and poor image quality of the segmented acetabular images can all easily lead to unreasonable or inaccurate selection of registration points, thus affecting the accuracy of subsequent registration.
[0022] To address the aforementioned issues, this application provides an embodiment that generates registration guide points to guide the surgeon in selecting registration points reasonably and accurately for subsequent registration operations, thereby greatly improving registration accuracy.
[0023] The technical solution of this application will be described below through specific embodiments.
[0024] Reference Figure 1 The diagram illustrates a method for generating acetabular registration points according to an embodiment of this application, which may specifically include the following steps: S101. Acquire the patient's hip bone image and perform acetabular identification in the hip bone image to obtain the identification result, the identification result including the image quality of the image region where the acetabular is located in the hip bone image.
[0025] It should be noted that this method can be applied to computer devices, meaning that the executing entity in this application embodiment can be a computer device. By executing the various steps of this method, the computer device can reasonably and accurately determine multiple registration points based on acetabular recognition of any hip bone image, for subsequent registration operations, thereby improving the accuracy of acetabular surgical registration. The aforementioned computer device can be an electronic device with computer-aided navigation-related functions, such as a desktop computer; this application embodiment does not limit the type of computer device.
[0026] In implementing this method, the computer device can first acquire images of the patient's hip bone and then identify the acetabulum in the hip bone images.
[0027] The hip bone image in this embodiment can be extracted from the image data of the patient's hip bone. The computer device can identify the acetabulum in the hip bone image through image recognition or processing techniques to obtain the corresponding recognition result.
[0028] In one possible implementation of this application, the identification result obtained by the computer device in identifying the acetabulum in the hip bone image can include two types: one is that the computer device can accurately identify the complete acetabulum from the hip bone image; the other is that the computer device cannot identify the acetabulum from the hip bone image. For example, as described above, if the bone boundary is blurred in the image and difficult to distinguish from the surrounding soft tissue; or if the patient has a complex acetabular fracture, such as a comminuted fracture, the normal anatomical contour of the acetabulum will be disrupted, resulting in displacement and overlap of fracture fragments, making it difficult to accurately identify the acetabulum in the hip bone image.
[0029] For an acetabulum that can be accurately identified from a hip bone image, there are two different scenarios: one is that the identified acetabular image is of good quality and can be used for subsequent registration operations; the other is that although the acetabulum can be identified from the hip bone image, the identified acetabular image is of poor quality and cannot be directly used for subsequent registration. The quality of the acetabular image can be characterized by its resolution. For example, if an identified acetabular image has a high resolution, such as a resolution greater than or equal to a predetermined first resolution, it can be considered to have good quality; conversely, if an identified acetabular image has a low resolution, such as a resolution less than the aforementioned predetermined first resolution, it can be considered to have poor quality.
[0030] In some scenarios, the quality of the acetabular image can also be judged based on whether the computer device can accurately identify the relevant landmarks in the acetabulum during subsequent registration-related operations. This application embodiment does not limit this.
[0031] In response to the various recognition results mentioned above, the embodiments of this application can be processed in a targeted manner. That is, for cases where the acetabular image cannot be recognized, or where the acetabular image can be recognized but the quality of the recognized acetabular image is poor, processing is performed separately to ensure that, in any case, the surgeon can easily determine a reasonable and accurate registration point.
[0032] S102. Based on the identification result of the acetabulum, determine multiple registration guide points, including multiple points located inside and / or outside the image area where the acetabulum is located.
[0033] In this embodiment, based on different acetabular recognition results, the computer device can determine multiple registration guide points in the hip bone image. These registration guide points can be used by the surgeon for reference when selecting registration points. Depending on the recognition results, the location of the registration guide points in the hip bone image varies. For example, the registration guide points can be located inside and / or outside the image region where the acetabulum is located in the hip bone image.
[0034] In one possible implementation of this application, if the computer device can accurately identify the acetabulum in the hip bone image, and the identified acetabulum image is of good quality, this means that based on the identified acetabulum image, the computer device can correctly identify the marker points in the image, which can then be used as registration points. Therefore, in this case, the computer device can determine multiple registration guide points both inside and outside the image region where the acetabulum is located.
[0035] In another possible implementation of this application, if the computer device can accurately identify the acetabulum in the hip image, but the quality of the identified acetabulum image is poor, this means that based on the identified acetabulum image, the computer device may not be able to correctly identify the landmark points in the area where the acetabulum is located. Therefore, in this case, the computer device can determine multiple registration guide points outside the image area where the acetabulum is located, and use the multiple guide points outside the acetabulum to guide the surgeon to select registration points, which can also achieve high-precision registration.
[0036] In another possible implementation of this application, if the computer device cannot accurately identify the acetabulum from the hip bone image, the surgeon cannot use the markers in the acetabulum for registration. Therefore, the computer device can first roughly determine the image area where the acetabulum is located, and then determine multiple registration guide points outside the image area. The multiple guide points around the hip bone are used to guide the surgeon to select registration points, thereby achieving subsequent high-precision registration.
[0037] It should be noted that the determination of registration guide points can be performed after point registration (i.e., the coarse registration stage). That is, for the acquired hip bone images, the computer equipment can first perform point registration, and then execute the various steps of this method to determine multiple registration guide points before area registration (i.e., the fine registration stage). These multiple registration guide points guide the surgeon to acquire registration points in characteristic, regional, and dispersed locations. The registration guide points are only used for guidance and do not participate in subsequent registration calculations.
[0038] S103. Based on the multiple registration guide points, generate multiple registration points, which are used to perform surface registration on the acetabulum.
[0039] As mentioned above, the multiple registration guide points in this embodiment are mainly used to guide the surgeon in acquiring registration points at characteristic, regionalized, and dispersed locations during the area registration (fine registration) process after point registration (coarse registration). The registration guide points do not participate in the calculation and are only used for guidance. Therefore, when multiple registration guide points are determined, the computer device can display these multiple registration guide points in the hip bone image, guiding the surgeon to select multiple registration points based on each registration guide point. In response to the selection of multiple registration points, the computer device can perform area registration (i.e., fine registration) using the selected multiple registration points.
[0040] In this embodiment, the computer device acquires an image of the patient's hip bone and can identify the acetabulum within that image, obtaining corresponding identification results. These results may include information such as the image region containing the acetabulum and the image quality of that region. Based on the acetabulum identification results, the computer device first determines multiple registration guide points. These guide points can be located inside and / or outside the image region containing the acetabulum. Thus, multiple registration points can be generated based on these guide points, allowing for planar registration of the acetabulum. Applying the method provided in this embodiment assists surgeons in rationally and accurately selecting registration points for high-precision registration under various imaging conditions, reducing the surgeon's workload during registration, ensuring the reliability of registration results, and ultimately improving the overall efficiency and safety of acetabular surgery.
[0041] Reference Figure 2 The diagram illustrates another method for generating acetabular registration points according to an embodiment of this application, which may specifically include the following steps: S201. Obtain the patient's hip bone image and perform acetabular identification in the hip bone image to obtain the identification result.
[0042] The execution subject in this embodiment can be the same as in the previous embodiment, that is, the execution subject in this embodiment is a computer device.
[0043] In this embodiment of the application, the patient's hip bone image can be obtained from the patient's preoperative medical imaging data. For example, the aforementioned medical imaging data can be CT image data, MRI image data, etc.
[0044] Specifically, the computer device can acquire a patient's medical image data and segment the hip bone image from that data. For example, the computer device can be configured with a segmentation algorithm that, after acquiring the patient's CT image data, can be used to segment the hip bone image from the CT image data for further processing.
[0045] In this embodiment of the application, further processing of the segmented hip bone image may involve identifying the acetabulum within the image to obtain a corresponding identification result. This identification result may include whether the acetabulum can be accurately identified from the hip bone image, and, if the acetabulum can be identified, the image quality of the image region containing the acetabulum in the hip bone image. This image quality can be characterized by image resolution. For example, a higher image resolution (greater than or equal to a certain set resolution threshold) indicates better image quality; conversely, a lower image resolution (less than a certain set resolution threshold) indicates worse image quality. In other words, the quality of the image is positively correlated with its resolution.
[0046] In one possible implementation of this application, an acetabular recognition model can be pre-trained. After segmenting the hip bone image from CT image data, the computer device can call the pre-trained acetabular recognition model to recognize the hip bone image and determine the image region where the acetabulum is located within the hip bone image. Specifically, the hip bone image can be used as input data for the acetabular recognition model, and the output data of the model can be the coordinates of relevant points in the image region where the acetabulum is located. For example, the acetabular recognition model can output the coordinates of various points constituting the edge of the acetabulum, and can also output some landmark points (anatomical landmarks) with specific anatomical significance in the acetabulum.
[0047] In this embodiment, the aforementioned acetabular recognition model can be trained based on pre-labeled acetabular sample data. For example, multiple sample data can be pre-collected, which may include hip bone image data including the acetabulum. Then, a doctor or an operator with the corresponding skills marks the location of the acetabulum and relevant anatomical landmarks within it in the sample data. A portion of the sample data can be used as a training set, and the remaining portion can be used as a validation set, thereby training a neural network model to obtain the aforementioned acetabular recognition model capable of acetabular recognition. By configuring this model in a computer device or in another device capable of communicating with the computer device, when an acetabular recognition task needs to be performed, the computer device calls the model to process the hip bone image and obtain the corresponding recognition result.
[0048] In one possible implementation of this application embodiment, during the training of the above-mentioned acetabular recognition model, an image enhancement module can be added to the model. The image enhancement module can be used to perform image enhancement processing on the image area where the acetabular is located before the acetabular recognition model outputs the acetabular recognition result, so as to improve the quality of the acetabular image as much as possible.
[0049] In this embodiment, if the acetabular recognition model can accurately identify the image region of the acetabulum in the hip bone image and the relevant anatomical landmarks in that image region, the quality of the currently identified acetabular image can be considered good (e.g., the image resolution of the image region where the acetabulum is located is greater than or equal to the first resolution). If the acetabular recognition model cannot accurately identify the image region of the acetabulum in the hip bone image (e.g., the output result of the acetabular recognition model is empty, or the relevant location points cannot be accurately located in the hip bone image based on the output result of the acetabular recognition model), or the anatomical landmarks in the region cannot be further identified based on the determined image region, the quality of the currently identified acetabular image can be considered poor (e.g., the image resolution of the image region where the acetabulum is located is less than the first resolution). The computer device can execute S202 or S203 to process different situations separately and generate multiple registration guide points at different positions in the hip bone image.
[0050] S202, if the image resolution of the image area where the acetabulum is located is less than the first resolution, generate multiple registration guide points outside the image area.
[0051] In this embodiment, if the image resolution of the image region where the acetabulum is located, as identified by the acetabular recognition model, is less than a first resolution, the quality of the currently identified acetabular image can be considered poor. In this case, multiple registration guide points can be generated outside the image region, and the surgeon can be guided to select registration points using the relatively better recognition results of the outer region.
[0052] In one possible implementation of this application, if the computer device fails to identify the acetabulum in the hip bone image after processing the input hip bone image using the acetabular recognition model, the computer device can generate manual operation prompts in the displayed hip bone image. These prompts instruct the surgeon to manually mark an image region in the hip bone image. This image region can be a region where the surgeon can visually observe the possible location of the acetabulum in the hip bone image. In response to the image region marked by the surgeon in the hip bone image according to the manual operation prompts, the computer device can generate multiple registration guide points outside the marked image region.
[0053] The following describes the specific process of generating multiple registration guide points outside the image region. In the example below, the image region can be the image region where the acetabulum is located in the hip bone image obtained by the computer device calling the acetabulum recognition model, or it can be the image region marked by the surgeon according to the manual operation prompts when the acetabulum recognition model fails to recognize the acetabulum.
[0054] like Figure 3 The diagram shown is a schematic representation of generating registration guide points according to an embodiment of this application. Figure 3 The illustrated process, when generating multiple registration guide points outside the image region where the acetabulum is located, can first determine a first guide region outside that image region. For example... Figure 3 As shown in (a) of the figure, an image region 301 containing the acetabulum and a first guide region 302 (represented by a dashed circle in the figure) outside the image region 301 are illustrated. Based on this, as... Figure 3 As shown in (b), multiple candidate guide points can be generated in the first guide region 302, for example... Figure 3 In (b), the points represented by hollow small circles are the candidate guide points. It should be noted that... Figure 3 The candidate guide points shown in (b) are only a portion of the candidate guide points.
[0055] In one possible implementation of this application embodiment, the computer device can generate multiple candidate guide points in the first guide region 302 by matching the currently segmented hip bone image with multiple reference hip bone images.
[0056] Specifically, the computer device can acquire multiple reference hip images, each of which is pre-labeled with multiple candidate guide points. These reference hip images can be acquired during the training of the aforementioned acetabular recognition model. On the one hand, by labeling the acetabulum in the hip images, they can be used as sample data to train the acetabular recognition model; on the other hand, multiple candidate guide points can also be labeled in the acquired hip images, and the labeled hip images can be used as reference images in the process of generating registration guide points.
[0057] The computer device can match the hip bone image with multiple reference hip bone images to obtain the similarity between the hip bone image and each reference hip bone image. In this way, multiple candidate guide points can be determined in the first guide region 302 outside the aforementioned image region 301 of the current hip bone image, based on multiple candidate guide points in the target reference hip bone image with a similarity greater than a preset value.
[0058] Then, the image region 301 can be fitted into a circle, and the circle can be divided into four quadrants with its center. For example... Figure 3 As shown in (c), point O represents the center of the fitted circle (not shown in the figure). Based on point O, an XY coordinate system can be constructed, thus dividing the image into four quadrants, i.e. Figure 3 Quadrants R1 to R4 are shown in (c) in the diagram.
[0059] The computer device can uniformly determine multiple registration guide points from multiple candidate guide points located in the intersection region of each quadrant and the aforementioned first guide region 302. For example, in Figure 3 The small solid black circles shown in (d) are the registration guide points.
[0060] In this embodiment of the application, 7 to 8 registration guide points can be generated in quadrants R1 to R4 respectively, thereby generating approximately 30 registration guide points outside the acetabular image region 301 for subsequent guidance.
[0061] For multiple candidate guide points that are not identified as registration guide points, the computer device can identify them as verification points. Each verification point can be used to verify the registration results after the acetabulum is registered.
[0062] S203. If the image resolution of the image region where the acetabulum is located is greater than or equal to the first resolution, multiple registration guide points are generated inside and outside the image region respectively.
[0063] In this embodiment, the acetabular image identified using the acetabular recognition model may also be of good quality. For example, the acetabular recognition model can accurately identify the acetabulum in the hip bone image, and the image resolution of the identified acetabular image is greater than or equal to a certain set resolution threshold, or the identified acetabular image can be directly used by a computer device for further processing. In this case, multiple anatomical landmarks can be determined both inside and outside the image region where the acetabulum is located, and then multiple registration guide points can be determined from these anatomical landmarks. The aforementioned anatomical landmarks can be locations in the hip bone image that have corresponding anatomical significance.
[0064] For example, suppose Figure 3 Image region 301 shown in (a) is the region where the acetabular image is located, as identified by the acetabular recognition model from the hip bone image. Since the image quality of this region is relatively good, the computer device can directly access the image region 301 (i.e., the content of the image region containing the acetabular) and the region outside of image region 301 (i.e.,...). Figure 3 Multiple registration guide points are generated for the area corresponding to the first guide region 302 in (a) of the diagram.
[0065] In one possible implementation of this application embodiment, the number of registration guide points located outside the image region can be greater than the number of registration guide points located inside the image region. For example, when the computer device can accurately identify the acetabulum image and the acetabulum image quality is good, 5 to 15 registration guide points can be generated inside the image region where the acetabulum is located, and 25 to 15 registration guide points can be generated outside the image region. The total number of registration guide points inside and outside the image region can be approximately 30.
[0066] The acetabular shape varies from patient to patient. For example, some patients have an acetabulum that is nearly spherical, nearly ellipsoidal, nearly planar, or irregularly shaped. If the acetabular features are not obvious, such as a spherical shape, acquiring registration guide points only within the acetabulum can easily lead to registration rotation. If registration guide points are acquired only outside the acetabulum, rotation may also occur during the registration process in some patients whose outside the acetabulum is nearly planar. Therefore, for cases with good acetabular image quality, multiple registration guide points can be generated simultaneously both inside and outside the acetabulum. By utilizing multiple registration guide points with diverse features, multiple less prone to rotation and more dispersed registration points can be generated, thereby further improving the registration accuracy.
[0067] S204. Display a plurality of the registration guide points in the image used for acetabular registration.
[0068] In this embodiment of the application, the computer device can display the generated multiple registration guide points in the image where acetabular registration is required, thereby guiding the surgeon to select multiple registration points that can be registered subsequently based on the displayed multiple guide points.
[0069] S205. In response to selecting multiple points in the image based on multiple registration guide points, as multiple registration points.
[0070] For example, based on the multiple registration guide points displayed, the surgeon can use a probe to select points in the image, and the selected points can be used as registration points for subsequent acetabular surface registration.
[0071] This application's embodiments, based on the obtained acetabular image, can selectively choose appropriate processing methods according to different image quality levels. By generating registration guide points, it guides the surgeon to rationally select registration points. For example, when the acetabular image quality is poor or the acetabulum cannot be identified, it guides the surgeon to acquire stable and clear anatomical points outside the acetabulum, bypassing the problem area, ensuring registration feasibility in extreme cases, and guaranteeing successful subsequent registration, avoiding the risk of being unable to continue surgery due to navigation failure. When the acetabular image quality is good, by combining registration points from both the inner and outer parts of the acetabulum, and utilizing local details and global frame information, it provides rich spatial constraints to the maximum extent, thereby ensuring high-precision registration results. Applying this application's embodiments can improve the overall efficiency and safety of acetabular surgery.
[0072] It should be noted that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.
[0073] To facilitate understanding, the following section provides a complete example to illustrate the process of acetabular registration using the registration points generated by the method provided in this application.
[0074] like Figure 4 The diagram shown is a schematic representation of an acetabular registration process provided in an embodiment of this application. According to... Figure 4 The procedure shown, to complete acetabular registration, mainly includes the following steps: (1) Image data acquisition: In this step, medical imaging equipment can be used to acquire the patient's imaging data before surgery. In this example, the imaging data can be CT image data.
[0075] (2) Acetate point registration: In this step, acetabular point registration can be performed based on the acquired CT image data, which is also known as coarse registration. Through point registration, a point registration matrix can be obtained, which is the registration result of the point registration (coarse registration) stage.
[0076] (3) Achilles tendon identification: In this step, a segmentation algorithm can be used to segment the hip bone image from the CT image data, and then the acetabulum can be identified within that hip bone image. In this example, the acetabulum identification results include three types: 1. The acetabulum is identifiable and the image quality is good; 2. The acetabulum is identifiable but the image quality is poor; 3. The acetabulum cannot be identified.
[0077] Different processing methods can be used for different recognition results.
[0078] (4) Generate registration guide points: In this step, different acetabular identification results obtained in the previous step can be processed specifically. Specifically: 1. For cases where the acetabulum is identifiable and the image quality is good, multiple registration guide points can be generated both inside and outside the acetabulum. For example, 5 to 15 guide points can be generated inside the acetabulum, and 25 to 15 guide points can be generated outside the acetabulum.
[0079] 2. The acetabulum can be identified, but the image quality is poor. Multiple registration guide points can be generated outside the acetabulum. For example, 30 guide points can be generated outside the acetabulum.
[0080] 3. If the acetabulum cannot be identified, a region can be manually marked in the image to represent the area where the acetabulum is located. Then, multiple registration guide points can be generated outside this region. For example, 30 guide points can be generated outside this region.
[0081] (5) Guide the generation of registration points: In this step, the surgeon can be guided to select points in the image based on the registration guide points generated in the previous step. Each selected point can then be used as a registration point for subsequent registration.
[0082] (6) Acetate surface registration: In this step, based on the registration points selected by the surgeon, and combined with the point registration matrix generated by point registration (coarse registration), surface registration (i.e. fine registration) can be performed to obtain the registration result.
[0083] Furthermore, based on the obtained registration results, registration verification points can be used to verify the surface registration results, further confirming the reliability of the registration results. Registration verification points can be generated simultaneously when generating registration guide points.
[0084] This completes the acetabular registration process.
[0085] Reference Figure 5 This diagram illustrates a acetabular registration point generation device according to an embodiment of this application. Specifically, it may include a hip bone image acquisition module 501, an acetabular recognition module 502, a registration guide point determination module 503, and a registration point generation module 504, wherein: Hip bone image acquisition module 501 is used to acquire hip bone images of the patient; The acetabular recognition module 502 is used to perform acetabular recognition in the hip bone image and obtain a recognition result, the recognition result including the image quality of the image region where the acetabular is located in the hip bone image; The registration guide point determination module 503 is used to determine multiple registration guide points based on the recognition result of the acetabulum. The registration guide points include multiple points located inside and / or outside the image area where the acetabulum is located. The registration point generation module 504 is used to generate multiple registration points based on multiple registration guide points, and the multiple registration points are used for surface registration of the acetabulum.
[0086] In one possible implementation of this application embodiment, the hip bone image acquisition module 501 may specifically be used for: The patient's medical imaging data is acquired, and the hip bone image is segmented from the medical imaging data, which includes CT image data.
[0087] In one possible implementation of this application embodiment, the acetabular recognition module 502 can specifically be used for: A pre-trained acetabular recognition model is invoked to identify the hip bone image and determine the image region where the acetabulum is located in the hip bone image; wherein, the acetabular recognition model is trained based on multiple pre-labeled acetabular sample data, and the acetabular recognition model includes an image enhancement module, which is used to perform image enhancement processing on the image region where the acetabulum is located before the acetabular recognition model outputs the recognition result of the acetabulum.
[0088] In one possible implementation of this application embodiment, the image quality is characterized by image resolution, and the registration guide point determination module 503 can specifically be used for: When the image resolution of the image region where the acetabulum is located is less than the first resolution, multiple registration guide points are generated outside the image region. When the image resolution of the image region where the acetabulum is located is greater than or equal to the first resolution, multiple registration guide points are generated inside and outside the image region, respectively; wherein, the number of registration guide points located outside the image region is greater than the number of registration guide points located inside the image region.
[0089] In one possible implementation of this application embodiment, the registration guide point determination module 503 may also be used for: A first guiding region is determined outside the image region, and multiple candidate guiding points are generated within the first guiding region; The image region is fitted into a circle, and the circle is divided into four quadrants with its center. From a plurality of candidate guide points located in the intersection region of each of the quadrants and the first guide region, a plurality of registration guide points are uniformly determined; Multiple verification points are determined from a plurality of candidate guide points that have not been identified as the registration guide point. These verification points are used to verify the registration results after the acetabulum is registered.
[0090] In this embodiment of the application, the registration guide point determination module 503 can also be used for: Acquire multiple reference hip images, each of which is pre-marked with multiple candidate guide points; The hip bone image is matched with multiple reference hip bone images to obtain the similarity between the hip bone image and each reference hip bone image; Based on multiple candidate guide points in the target reference hip image with a similarity greater than a preset value, multiple candidate guide points are determined in a first guide region outside the image region.
[0091] In another possible implementation of this application embodiment, the registration guide point determination module 503 can also be used for: Multiple anatomical landmarks are determined both inside and outside the image region; Multiple registration guide points are determined from the multiple anatomical landmarks.
[0092] In another possible implementation of this application embodiment, the registration guide point determination module 503 can also be used for: If the acetabulum cannot be identified in the hip bone image, a manual operation prompt message is generated; In response to the image region marked in the hip bone image for the manual operation prompt information, a plurality of registration guide points are generated outside the marked image region.
[0093] In this embodiment of the application, the registration point generation module 504 can specifically be used for: Multiple registration guide points are shown in the image used for acetabular registration; In response to multiple points selected in the image based on multiple registration guide points, as multiple registration points.
[0094] This application provides an acetabular registration point generation device, which can be the computer device or a module or unit with corresponding functions in the aforementioned embodiments. Using this device, the steps in the aforementioned method embodiments can be implemented.
[0095] As the apparatus embodiments are basically similar to the method embodiments, they are described in a relatively simple manner. For relevant details, please refer to the description in the method embodiment section.
[0096] Reference Figure 6 The diagram illustrates a computer device provided in an embodiment of this application. Figure 6 As shown, the computer device 600 in this embodiment includes: a processor 610, a memory 620, and a computer program 621 stored in the memory 620 and executable on the processor 610. When the processor 610 executes the computer program 621, it implements the steps in the various embodiments of the acetabular registration point generation method described above, for example... Figure 1 Steps S101 to S103 shown, or Figure 2 The steps S201 to S205 are shown. Alternatively, when the processor 610 executes the computer program 621, it implements the functions of each module / unit in the above-described device embodiments, for example... Figure 5 The functions of modules 501 to 504 are shown.
[0097] For example, the computer program 621 can be divided into one or more modules / units, which are stored in the memory 620 and executed by the processor 610 to complete this application. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, which can be used to describe the execution process of the computer program 621 in the computer device 600. For example, the computer program 621 can be divided into a hip bone image acquisition module, an acetabular recognition module, a registration guide point determination module, and a registration point generation module, with the specific functions of each module as follows: The hip bone image acquisition module is used to acquire images of the patient's hip bone. An acetabular recognition module is used to perform acetabular recognition in the hip bone image and obtain a recognition result, the recognition result including the image quality of the image region where the acetabular is located in the hip bone image; The registration guide point determination module is used to determine multiple registration guide points based on the recognition result of the acetabulum. The registration guide points include multiple points located inside and / or outside the image area where the acetabulum is located. The registration point generation module is used to generate multiple registration points based on multiple registration guide points, and the multiple registration points are used for surface registration of the acetabulum.
[0098] The computer device 600 may be an electronic device capable of implementing the various steps or corresponding functions in the foregoing method embodiments. The computer device 600 may be a desktop computer, a cloud server, or other similar device. The computer device 600 may include, but is not limited to, a processor 610 and a memory 620. Those skilled in the art will understand that... Figure 6 This is merely one example of computer device 600 and does not constitute a limitation on computer device 600. It may include more or fewer components than shown, or combine certain components, or different components. For example, computer device 600 may also include input / output devices, network access devices, buses, etc.
[0099] The processor 610 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.
[0100] The memory 620 can be an internal storage unit of the computer device 600, such as a hard disk or RAM of the computer device 600. The memory 620 can also be an external storage device of the computer device 600, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, etc., equipped on the computer device 600. Furthermore, the memory 620 can include both internal and external storage units of the computer device 600. The memory 620 is used to store the computer program 621 and other programs and data required by the computer device 600. The memory 620 can also be used to temporarily store data that has been output or will be output.
[0101] This application also discloses a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it implements the methods described in the foregoing embodiments.
[0102] This application also discloses a computer-readable storage medium storing a computer program that, when executed by a computer, implements the methods described in the foregoing embodiments.
[0103] This application also discloses a computer program product, including a computer program that, when run on a computer, causes the computer to perform the methods described in the foregoing embodiments.
[0104] The embodiments described above are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.
Claims
1. A method for generating acetabular registration points, characterized in that, include: Acquire a patient's hip bone image and perform acetabular identification in the hip bone image to obtain an identification result, the identification result including the image quality of the image region where the acetabular is located in the hip bone image; Based on the identification result of the acetabulum, multiple registration guide points are determined, including multiple points located inside and / or outside the image region where the acetabulum is located; Based on the multiple registration guide points, multiple registration points are generated, and the multiple registration points are used for surface registration of the acetabulum.
2. The method according to claim 1, characterized in that, The process of acquiring the patient's hip bone image and performing acetabular identification within the hip bone image to obtain the identification result includes: Acquire the patient's medical imaging data, and segment the hip bone image from the medical imaging data, which includes CT image data; A pre-trained acetabular recognition model is invoked to identify the hip bone image and determine the image region where the acetabulum is located in the hip bone image; wherein, the acetabular recognition model is trained based on multiple pre-labeled acetabular sample data, and the acetabular recognition model includes an image enhancement module, which is used to perform image enhancement processing on the image region where the acetabulum is located before the acetabular recognition model outputs the recognition result of the acetabulum.
3. The method according to claim 1 or 2, characterized in that, The image quality is characterized by image resolution, and based on the identification result of the acetabulum, multiple registration guide points are determined, including: When the image resolution of the image region where the acetabulum is located is less than the first resolution, multiple registration guide points are generated outside the image region. When the image resolution of the image region where the acetabulum is located is greater than or equal to the first resolution, multiple registration guide points are generated inside and outside the image region, respectively; wherein, the number of registration guide points located outside the image region is greater than the number of registration guide points located inside the image region.
4. The method according to claim 3, characterized in that, The generation of multiple registration guide points outside the image region includes: A first guiding region is determined outside the image region, and multiple candidate guiding points are generated within the first guiding region; The image region is fitted into a circle, and the circle is divided into four quadrants with its center. From a plurality of candidate guide points located in the intersection region of each of the quadrants and the first guide region, a plurality of registration guide points are uniformly determined; Multiple verification points are determined from a plurality of candidate guide points that have not been identified as the registration guide point. These verification points are used to verify the registration results after the acetabulum is registered.
5. The method according to claim 4, characterized in that, The step of generating multiple candidate guide points in the first guide region includes: Acquire multiple reference hip images, each of which is pre-marked with multiple candidate guide points; The hip bone image is matched with multiple reference hip bone images to obtain the similarity between the hip bone image and each reference hip bone image; Based on multiple candidate guide points in the target reference hip image with a similarity greater than a preset value, multiple candidate guide points are determined in a first guide region outside the image region.
6. The method according to claim 3, characterized in that, The process involves generating multiple registration guide points both inside and outside the image region, including: Multiple anatomical landmarks are determined both inside and outside the image region; Multiple registration guide points are determined from the multiple anatomical landmarks.
7. The method according to claim 1, characterized in that, Also includes: If the acetabulum cannot be identified in the hip bone image, a manual operation prompt message is generated; In response to the image region marked in the hip bone image for the manual operation prompt information, a plurality of registration guide points are generated outside the marked image region.
8. The method according to any one of claims 1, 2, or 4 to 7, characterized in that, The generation of multiple registration points based on the multiple registration guidance points includes: Multiple registration guide points are shown in the image used for acetabular registration; In response to multiple points selected in the image based on multiple registration guide points, as multiple registration points.
9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it causes the computer device to implement the method as described in any one of claims 1 to 8.
10. A computer program product, comprising a computer program, characterized in that, When the computer program is run, the method as described in any one of claims 1 to 8 is performed.