Femoral head center point identification method, apparatus, device, and computer program product

Abstract

Embodiments of the present application are applicable to the field of computer-aided medical technology and image processing technology, and provide a femoral head center point identification method, device, equipment and computer program product, the method comprising: acquiring femoral head image data to be identified; rendering the femoral head image data to obtain a first femoral head image and a second femoral head image; calling a femoral head center identifier to process the first femoral head image and the second femoral head image, obtaining two sets of two-dimensional coordinates output by the femoral head center identifier, the femoral head center identifier being trained based on a plurality of two-dimensional femoral head sample images labeled with femoral head center points; and determining the femoral head center point in the femoral head image data according to the two sets of two-dimensional coordinates. By using the above method, the femoral head center point can be quickly and accurately determined for any type of femoral head image data.

Description

Technical Field

[0001] This application belongs to the fields of computer-aided medical technology and image processing technology, and in particular relates to a method, device, equipment and computer program product for identifying the center point of the femoral head. Background Technology

[0002] Identifying joint landmarks in the human body has important applications in medical image analysis, orthopedic surgical planning, and auxiliary diagnosis. For example, the femoral head center point is a key reference point for preoperative planning, biomechanical analysis, and diagnosis of related diseases in hip replacement surgery, and is of great significance in orthopedic clinical diagnosis and surgical planning. Traditional methods for identifying the femoral head center point mainly rely on the doctor's experience and manual measurement. This method is not only time-consuming and labor-intensive, but also inefficient, and is susceptible to the influence of the doctor's subjective factors, making it difficult to guarantee the accuracy of the identification results.

[0003] With the development of medical imaging technology, the industry has begun to utilize computer equipment incorporating image processing techniques to assist in the automatic identification of human landmarks such as the femoral head center point. However, the efficiency of automatic identification of the femoral head center point depends on computer resources, especially for three-dimensional femoral head image data, where the efficiency of automatic computer identification still needs improvement. Summary of the Invention

[0004] In view of this, embodiments of this application provide a method, apparatus, device, and computer program product for identifying the center point of the femoral head, in order to improve the efficiency and accuracy of identifying the center point of the femoral head.

[0005] The first aspect of this application provides a method for identifying the center point of the femoral head, including:

[0006] Acquire the femoral head image data to be identified, wherein the femoral head image data is two-dimensional image data or three-dimensional image data;

[0007] The femoral head image data is rendered to obtain a first femoral head image and a second femoral head image, both of which are two-dimensional images.

[0008] The femoral head center recognizer is invoked to process the first femoral head image and the second femoral head image to obtain two sets of two-dimensional coordinates output by the femoral head center recognizer. The femoral head center recognizer is trained based on multiple two-dimensional femoral head sample images labeled with the femoral head center point.

[0009] The center point of the femoral head in the femoral head image data is determined based on the two sets of two-dimensional coordinates.

[0010] Optionally, rendering the femoral head image data to obtain a first femoral head image and a second femoral head image includes:

[0011] The femoral head image data is rendered into a three-dimensional space;

[0012] Determine the first-person and second-person perspectives;

[0013] Image acquisition is performed on the femoral head image data in the three-dimensional space based on the first viewpoint and the second viewpoint respectively to obtain the first femoral head image and the second femoral head image.

[0014] Optionally, determining the first viewpoint and the second viewpoint includes:

[0015] Construct a three-dimensional coordinate system corresponding to the three-dimensional space;

[0016] A target surface in the three-dimensional space is determined, and a first viewpoint is determined in the direction facing the target surface. The target surface is a plane perpendicular to any axis in the three-dimensional coordinate system.

[0017] Determine a second perspective that is orthogonal to the first perspective.

[0018] Optionally, the step of acquiring images of the femoral head image data in the three-dimensional space based on the first viewpoint and the second viewpoint respectively to obtain a first femoral head image and a second femoral head image includes:

[0019] Based on the first perspective, the operation of acquiring the femoral head image data using a first camera at the first perspective position is simulated to generate a first femoral head image acquired by the first camera.

[0020] Based on the second perspective, the operation of acquiring the femoral head image data using a second camera at the second perspective position is simulated to generate a second femoral head image acquired by the second camera.

[0021] Optionally, the two sets of two-dimensional coordinates include a first coordinate and a second coordinate. The first coordinate is the coordinate of the first femoral head center point output by the femoral head center identifier for the first femoral head image, and the second coordinate is the coordinate of the second femoral head center point output by the femoral head center identifier for the second femoral head image. The first viewpoint and the second viewpoint are orthogonal viewpoints. Determining the femoral head center point in the femoral head image data based on the two sets of two-dimensional coordinates includes:

[0022] Determine a first line from the first viewpoint to the center point of the first femoral head, and a second line from the second viewpoint to the center point of the second femoral head, wherein the extensions of the first line and the second line intersect at a point in space.

[0023] Project the intersection point of the extension of the first line and the extension of the second line onto the three-dimensional space corresponding to the femoral head image data;

[0024] The projection point in the three-dimensional space is identified as the center point of the femoral head in the femoral head image data.

[0025] Optionally, projecting the intersection point of the extensions of the first and second connecting lines onto the space corresponding to the femoral head image data includes:

[0026] Determine the spatial matrix and calibration matrix corresponding to the first viewpoint, wherein the first viewpoint is facing the target surface of the femoral head image data, and the target surface is a plane perpendicular to any axis in the coordinate system of the three-dimensional space;

[0027] Based on the spatial matrix and the calibration matrix, matrix calculations are performed on the intersection points to obtain the coordinates of the femoral head center point in the three-dimensional space corresponding to the femoral head image data.

[0028] Optionally, the two-dimensional femoral head sample image used to train the femoral head center recognizer includes multiple two-dimensional images obtained by reconstructing multiple femoral head sample image data from multiple viewpoints, wherein the number of two-dimensional femoral head sample images reconstructed based on any of the femoral head sample image data is equal to the number of multiple viewpoints.

[0029] A second aspect of this application provides a femoral head center point identification device, comprising:

[0030] The acquisition module is used to acquire the femoral head image data to be identified, wherein the femoral head image data is two-dimensional image data or three-dimensional image data;

[0031] The rendering module is used to render the femoral head image data to obtain a first femoral head image and a second femoral head image, both of which are two-dimensional images.

[0032] The processing module is used to call the femoral head center recognizer to process the first femoral head image and the second femoral head image to obtain two sets of two-dimensional coordinates output by the femoral head center recognizer. The femoral head center recognizer is trained based on multiple two-dimensional femoral head sample images labeled with the femoral head center point.

[0033] The determination module is used to determine the center point of the femoral head in the femoral head image data based on the two sets of two-dimensional coordinates.

[0034] 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 as described in any of the first aspects above.

[0035] 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 any of the first aspects above.

[0036] A fifth aspect of this application provides a computer program product, including a computer program that, when the computer program is run, causes the method described in any of the first aspects above to be executed.

[0037] Compared with the prior art, the embodiments of this application have the following beneficial effects:

[0038] In this embodiment, a femoral head center recognizer, pre-trained to automatically process and identify two-dimensional images and output the coordinates of the femoral head center point in the image, can render the femoral head image data into three-dimensional space after acquiring the femoral head image data to be identified. This allows for the acquisition of first and second femoral head images from different viewpoints. By calling the pre-trained femoral head center recognizer to process the first and second femoral head images, two sets of two-dimensional coordinates output by the recognizer can be obtained. A computer device can then calculate the femoral head center point in the femoral head image data based on these two sets of coordinates. This embodiment improves the efficiency and accuracy of femoral head center point recognition by training a femoral head center recognizer to automatically process femoral head image data and identify the femoral head center point. Furthermore, in this embodiment, the input data for the femoral head center identifier is a two-dimensional image. If the femoral head image data to be identified is three-dimensional, identification can be performed simply by acquiring two-dimensional images from different perspectives. Compared to other existing machine learning models that require direct model processing of three-dimensional image data, this method can improve model processing speed by more than 100 times, significantly enhancing the efficiency of femoral head center point identification. In addition, this embodiment can process any type of image data, whether two-dimensional or three-dimensional. The identification method provided in this embodiment can yield accurate identification results. Attached Figure Description

[0039] 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.

[0040] Figure 1 This is a schematic diagram of a method for identifying the center point of the femoral head provided in an embodiment of this application;

[0041] Figure 2 This is a schematic diagram of another method for identifying the center point of the femoral head provided in an embodiment of this application;

[0042] Figure 3 This is a schematic diagram of a possible implementation of S202 in the femoral head center point identification method provided in the embodiments of this application;

[0043] Figure 4 This is a schematic diagram of a femoral head center point identification process provided in an embodiment of this application;

[0044] Figure 5 This is a schematic diagram of a femoral head center point identification device provided in an embodiment of this application;

[0045] Figure 6 This is a schematic diagram of a computer device provided in an embodiment of this application. Detailed Implementation

[0046] 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.

[0047] The technical solution of this application will be described below through specific embodiments.

[0048] Reference Figure 1 The diagram illustrates a method for identifying the center point of the femoral head according to an embodiment of this application, which may specifically include the following steps:

[0049] S101. Obtain the femoral head image data to be identified, wherein the femoral head image data is two-dimensional image data or three-dimensional image data.

[0050] It should be noted that this method can be applied to computer devices, that is, the execution subject of this application embodiment is a computer device. The computer device can automatically identify the femoral head center point in the femoral head image data by executing each step of this method, thereby improving the identification efficiency and accuracy of the femoral head center point.

[0051] The femoral head imaging data to be identified can be obtained from imaging scans of the relevant areas of the patient's femoral head. This scanning process can be performed preoperatively or include intraoperative scans. Preoperative femoral head imaging data can be used for preoperative planning, while intraoperative femoral head imaging data can provide a reference for the surgeon during the intraoperative phase.

[0052] In this application embodiment, the femoral head imaging data to be identified can be of any type. For example, the femoral head imaging data can be two-dimensional or three-dimensional. Exemplarily, two-dimensional imaging data can include X-ray images, etc., and three-dimensional imaging data can include computed tomography (CT) data, magnetic resonance imaging (MRI) data, etc. This application embodiment does not limit the type of femoral head imaging data.

[0053] Any type of femoral head image data can be acquired or scanned using appropriate imaging equipment. For example, the X-ray image mentioned above can be acquired using an X-ray machine, and CT data or MRI data can be acquired using CT equipment or MRI equipment, respectively. The image data containing the femoral head region acquired by the imaging equipment is the femoral head image data in the embodiments of this application. The imaging equipment can transmit the femoral head image data to a computer device, which will process it and automatically identify the center point of the femoral head.

[0054] S102. Render the femoral head image data to obtain a first femoral head image and a second femoral head image, both of which are two-dimensional images.

[0055] In this embodiment of the application, rendering of femoral head image data can refer to the process by which a computer device renders the femoral head image data into a three-dimensional space after receiving it. Whether the image data is two-dimensional or three-dimensional, the computer device can render it into a three-dimensional space.

[0056] The purpose of rendering femoral head image data into three-dimensional space is to facilitate image acquisition based on the rendered three-dimensional image data, thus obtaining a two-dimensional image. In other words, rendering allows the subsequent recognition process to be converted into a processing procedure for two-dimensional images, reducing the computer resources required for recognition and facilitating the deployment of this method across a wider range of devices. Simultaneously, since the subsequent recognition process is performed on two-dimensional images, it also improves the efficiency of recognizing the femoral head center point compared to the processing of three-dimensional data by computer equipment.

[0057] In this embodiment of the application, for femoral head image data that is originally three-dimensional image data, it can be directly rendered into three-dimensional space using standard processing methods; for femoral head image data that is two-dimensional image data, such as two-dimensional images, it can be rendered into three-dimensional space using different window widths and window levels. This embodiment of the application does not limit this.

[0058] After rendering the femoral head image data into a three-dimensional space, the computer device can acquire a first femoral head image and a second femoral head image based on the image data in the three-dimensional space. Both the first femoral head image and the second femoral head image are two-dimensional images.

[0059] In one example, two cameras can be used to capture images of the femoral head rendered in three-dimensional space, resulting in a first femoral head image and a second femoral head image. Alternatively, a computer device can simulate the acquisition of images of the femoral head from different perspectives using cameras to generate a first femoral head image and a second femoral head image. This application does not limit the scope of the embodiments.

[0060] S103. The femoral head center recognizer is invoked to process the first femoral head image and the second femoral head image to obtain two sets of two-dimensional coordinates output by the femoral head center recognizer. The femoral head center recognizer is trained based on multiple two-dimensional femoral head sample images labeled with the femoral head center point.

[0061] In this embodiment, the femoral head center identifier can be a pre-trained model capable of processing two-dimensional images and automatically identifying the femoral head center point in the image. The training of the aforementioned femoral head center identifier can be achieved using multiple two-dimensional femoral head sample images labeled with the femoral head center point.

[0062] In one possible implementation of this application, the two-dimensional femoral head sample images used to train the femoral head center recognizer may include multiple image sets. Each image set contains two sample images from different perspectives of the same data, and the position of the femoral head center point is pre-marked in both sample images. For example, the coordinates of the femoral head center point in the sample image. In this way, multiple pairs of data {sample image, femoral head center point coordinates} can be formed.

[0063] The computer device can be trained on the multiple pairs of data {sample image, coordinates of the femoral head center point} formed above to obtain a femoral head center recognizer.

[0064] In one possible implementation of this application, a femoral head center recognizer can be obtained by training a model on a two-dimensional femoral head sample image based on relevant object detection algorithms or models in the field of computer vision. For example, the femoral head center recognizer can be trained based on a Gaussian heatmap or various versions of the YOLO algorithm. Taking a Gaussian heatmap as an example, a heatmap can be trained for each point in the sample image. Through image segmentation and regression processing, the femoral head center point in the image is determined, and then the convergence of the model is judged by combining it with the actual femoral head center point in the labeled image. If the corresponding convergence conditions are met, the output model is used as the femoral head center recognizer for subsequent automatic identification of the femoral head center point.

[0065] The aforementioned femoral head center identifier can be deployed in a computer device or in other devices or servers connected to that computer device, such as in the cloud. After the computer device completes the rendering of the femoral head image data and obtains the first and second femoral head images, the femoral head center identifier can be called to process the two two-dimensional images and output the two-dimensional coordinates (u, v) of the femoral head center point.

[0066] In this embodiment, the femoral head center recognizer can process the input first femoral head image and the second femoral head image respectively, and output the two-dimensional coordinates of the femoral head center point. That is, for the first femoral head image, the femoral head center recognizer can output a set of two-dimensional coordinates (u1, v1); for the second femoral head image, the femoral head center recognizer can also output a set of two-dimensional coordinates (u2, v2). In this way, two sets of two-dimensional coordinates are obtained.

[0067] S104. Determine the center point of the femoral head in the femoral head image data based on the two sets of two-dimensional coordinates.

[0068] The computer device can calculate the coordinates of the femoral head center point in the femoral head image data based on the two sets of two-dimensional coordinates output by the femoral head center identifier.

[0069] In this embodiment, since the first femoral head image and the second femoral head image are two-dimensional images acquired from the same femoral head image data from different perspectives, the two sets of two-dimensional coordinates output by the femoral head center recognizer actually represent the coordinates of the same point from different perspectives. Based on the above principle, a position point can be determined in the rendered three-dimensional space according to the two sets of two-dimensional coordinates. This position point represents the center point of the femoral head in the femoral head image data to be identified.

[0070] In this embodiment, a femoral head center recognizer, pre-trained to automatically process and identify two-dimensional images and output the coordinates of the femoral head center point in the image, can render the femoral head image data into three-dimensional space after acquiring the femoral head image data to be identified. This allows for the acquisition of first and second femoral head images from different viewpoints. By calling the pre-trained femoral head center recognizer to process the first and second femoral head images, two sets of two-dimensional coordinates output by the recognizer can be obtained. A computer device can then calculate the femoral head center point in the femoral head image data based on these two sets of coordinates. This embodiment improves the efficiency and accuracy of femoral head center point recognition by training a femoral head center recognizer to automatically process femoral head image data and identify the femoral head center point. Furthermore, in this embodiment, the input data for the femoral head center identifier is a two-dimensional image. If the femoral head image data to be identified is three-dimensional, identification can be performed simply by acquiring two-dimensional images from different perspectives. Compared to other existing machine learning models that require direct model processing of three-dimensional image data, this method can improve model processing speed by more than 100 times, significantly enhancing the efficiency of femoral head center point identification. In addition, this embodiment can process any type of image data, whether two-dimensional or three-dimensional. The identification method provided in this embodiment can yield accurate identification results.

[0071] Reference Figure 2 This diagram illustrates another method for identifying the center point of the femoral head provided in an embodiment of this application, which may specifically include the following steps:

[0072] S201. Obtain the femoral head image data to be identified, and render the femoral head image data into a three-dimensional space.

[0073] The execution subject of this application embodiment is a computer device. The computer device can communicate with an imaging device. After the imaging device scans the patient's femoral head region and obtains femoral head image data, the imaging device can transmit the femoral head image data to the computer device. The femoral head image data in this application embodiment can refer to medical imaging data containing the patient's femoral head region, such as X-ray images, CT data, or MRI data.

[0074] For the imaging data of the femoral head to be identified, computer equipment can render it into three-dimensional space. That is:

[0075] {X-Ray, CT, MRI}→{S3}, where X-Ray represents X-ray images, CT represents CT data, MRI represents MRI data, and S3 represents three-dimensional space.

[0076] It should be noted that the femoral head center point recognition method provided in this application embodiment can process and recognize any type of femoral head image data, and output the coordinates of the femoral head center point. Before processing, whether it is two-dimensional image data or three-dimensional image data, computer equipment can render it into three-dimensional space.

[0077] S202. Image acquisition is performed on the femoral head image data in the three-dimensional space based on the first viewpoint and the second viewpoint respectively to obtain the first femoral head image and the second femoral head image.

[0078] In this embodiment, after rendering the femoral head image data to be identified into a three-dimensional space, the computer device can determine a first viewpoint and a second viewpoint, and obtain a first femoral head image and a second femoral head image based on the first and second viewpoints. This process is similar to acquiring images of the femoral head image data rendered into three-dimensional space from the first and second viewpoints, respectively. Therefore, the first femoral head image can refer to the image acquired from the first viewpoint, and the second femoral head image can refer to the image acquired from the second viewpoint. Both the first and second femoral head images are two-dimensional images.

[0079] In one possible implementation of the embodiments of this application, such as Figure 3 As shown, S202 may specifically include the following sub-steps S2021-S2023:

[0080] S2021. Construct a three-dimensional coordinate system corresponding to the three-dimensional space.

[0081] S2022. Determine the target surface in the three-dimensional space and determine the first viewpoint in the direction facing the target surface, wherein the target surface is a plane perpendicular to any axis in the three-dimensional coordinate system.

[0082] S2023. Determine a second perspective that is orthogonal to the first perspective.

[0083] Specifically, the computer device can first construct a three-dimensional coordinate system based on a point in three-dimensional space, with that point as the origin. This three-dimensional coordinate system can be represented as XYZ, i.e., the three-dimensional coordinate system XYZ. Then, a target surface is determined in the three-dimensional space. This target surface can be a plane perpendicular to any axis in the three-dimensional coordinate system XYZ. For example, the target surface can be the XY plane in the three-dimensional coordinate system XYZ, where the XY plane is perpendicular to the Z-axis of the three-dimensional coordinate system XYZ; or, the target surface can be the YZ plane in the three-dimensional coordinate system XYZ, where the YZ plane is perpendicular to the X-axis of the three-dimensional coordinate system XYZ; or, the target surface can be the XZ plane in the three-dimensional coordinate system XYZ. This application embodiment does not limit this.

[0084] Based on this, a first perspective and a second perspective can be determined. For example, the first perspective can be determined by the direction facing the target surface. For instance, when the target surface is the XY plane in the three-dimensional coordinate system XYZ, the direction of the first perspective can be facing that XY plane.

[0085] In this embodiment, the first viewpoint and the second viewpoint can be orthogonal. Therefore, when the first viewpoint is determined based on the target surface in three-dimensional space, the direction of the second viewpoint can be determined according to the orthogonality.

[0086] Then, image acquisition can be performed on the femoral head image data in three-dimensional space based on the first and second perspectives respectively, to obtain the first femoral head image and the second femoral head image.

[0087] Specifically, the computer device can, based on a first perspective, simulate the operation of acquiring image data of the femoral head using a first camera at the first perspective position, generating a first femoral head image acquired by the first camera. Similarly, the computer device can, based on a second perspective, simulate the operation of acquiring image data of the femoral head using a second camera at the second perspective position, generating a second femoral head image acquired by the second camera.

[0088] Thus, for any type of femoral head image data to be identified, the aforementioned steps can yield two femoral head images: a first femoral head image and a second femoral head image. This process can be represented as: {X-Ray, CT, MRI} → {S3} → {P1, P2}, where P1 represents the first femoral head image and P2 represents the second femoral head image.

[0089] S203. The femoral head center recognizer is invoked to process the first femoral head image and the second femoral head image to obtain two sets of two-dimensional coordinates output by the femoral head center recognizer.

[0090] The femoral head center recognizer can be pre-trained. For details on how to train the femoral head center recognizer, please refer to the description in section S103 of the previous embodiment, which will not be repeated here.

[0091] The femoral head center recognizer can process and recognize the first and second femoral head images separately, outputting the corresponding femoral head center point coordinates (u, v). That is: (u, v) = model(P), where model represents the trained femoral head center recognizer, and P represents the first or second femoral head image. The entire process can also be represented as: {X-Ray, CT, MRI} → {S3} → {P1, P2} → {(u, v)}.

[0092] Furthermore, for the first femoral head image, the femoral head center recognizer can output a set of two-dimensional coordinates, namely the first coordinates (u1, v1); for the second femoral head image, the femoral head center recognizer can also output a set of two-dimensional coordinates, namely the second coordinates (u2, v2). That is, for any femoral head image, the femoral head center recognizer can output a set of two-dimensional coordinates. In this way, a correspondence can be established between the image and the coordinates of the femoral head center point: {P: (u, v)}, where P represents any femoral head image, and (u, v) represents the two-dimensional coordinates output by the femoral head center recognizer based on that image P, which are the coordinates of the femoral head center point in image P.

[0093] Thus, after processing the aforementioned steps, the computer device can obtain two sets of two-dimensional coordinates for the femoral head image data to be identified. Based on these two sets of two-dimensional coordinates, the computer device can determine the center point of the femoral head.

[0094] S204. Determine a first line from the first viewpoint position to the center point of the first femoral head, and a second line from the second viewpoint position to the center point of the second femoral head.

[0095] In this embodiment of the application, when determining the femoral head center point in the femoral head image data based on two sets of two-dimensional coordinates, the first identified femoral head center point can be connected to the first viewpoint position to obtain a first connecting line; the second identified femoral head center point can be connected to the second viewpoint position to obtain a second connecting line. Here, the first femoral head center point can refer to the position point corresponding to the first coordinate (u1, v1) in the aforementioned example, the second femoral head center point can refer to the position point corresponding to the second coordinate (u2, v2) in the aforementioned example, the first viewpoint position can refer to the position point where the first camera is located, and the second viewpoint position can refer to the position point where the second camera is located.

[0096] S205. Project the intersection point of the extension of the first line and the extension of the second line onto the three-dimensional space corresponding to the femoral head image data.

[0097] In this embodiment, the first and second perspectives are orthogonal. Therefore, the extensions of the first and second connecting lines will intersect at a point. By reprojecting this point into three-dimensional space, the center point of the femoral head in the femoral head image data can be obtained.

[0098] Specifically, the intersection points can be projected into three-dimensional space using either a first-view or a second-view perspective. For example, taking the first-view perspective as an example, the spatial matrix and calibration matrix corresponding to the first-view perspective can be determined. Then, matrix calculations can be performed on the intersection points based on the spatial matrix and the calibration matrix to obtain the coordinates of the femoral head center point in the three-dimensional space corresponding to the femoral head image data.

[0099] S206. Identify the projection point in the three-dimensional space as the center point of the femoral head in the femoral head image data.

[0100] 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.

[0101] To facilitate understanding, a complete example will be used to introduce the femoral head center point identification method provided in the embodiments of this application.

[0102] like Figure 4 The diagram shown is a schematic representation of a femoral head center point identification process provided in an embodiment of this application. Figure 4 The process shown mainly includes steps such as recognizer training, image rendering, and center point recognition, which are detailed below:

[0103] (1) Recognizer training

[0104] Right now Figure 4The step shown in S401. In this step, the femoral head center recognizer can be trained using sample images.

[0105] In this embodiment, the sample images used to train the femoral head center recognizer are labeled two-dimensional femoral head sample images. These two-dimensional femoral head sample images include multiple two-dimensional images obtained by reconstructing multiple femoral head sample image data from multiple viewpoints. The number of two-dimensional femoral head sample images reconstructed based on any femoral head sample image data is equal to the number of viewpoints. Thus, when the sample image data is limited, the number of sample images can be increased by increasing the viewpoint angle during reconstruction, thereby enabling faster model convergence and enhancing model stability.

[0106] During the process of acquiring sample images, sample images of any data type can be acquired, such as two-dimensional image data and three-dimensional image data, which can be used in the training of the recognizer in this method.

[0107] Specifically, for any type of sample image data, it can first be rendered into three-dimensional space, and two-dimensional images from different perspectives can be acquired. By annotating the femoral head center point of each two-dimensional image, pairs of data can be obtained, each pair of data including a two-dimensional image and the coordinate information of the femoral head center point in that two-dimensional image.

[0108] A large amount of paired data can be divided into training, testing, and validation sets for training a femoral head center recognizer. This femoral head center recognizer can be trained based on Gaussian heatmaps or YOLO models. Subsequently, the femoral head center recognizer can be used to identify the corresponding femoral head center point from femoral head image data.

[0109] (2) Image rendering

[0110] Right now Figure 4 Step S402 is shown in the diagram. In this step, a first femoral head image and a second femoral head image can be rendered from the femoral head image data to be identified for subsequent identification.

[0111] In this embodiment, since the input data of the femoral head center identifier is a two-dimensional image, it is necessary to render any type of femoral head image data to be identified into three-dimensional space and acquire two-dimensional images from two orthogonal perspectives. For example, a first femoral head image from a first perspective and a second femoral head image from a second perspective are acquired.

[0112] (3) Center point identification

[0113] Right now Figure 4The steps are shown in S403. In this step, a pre-trained femoral head center recognizer can be called to process the two-dimensional image and output two sets of two-dimensional coordinates; the femoral head center point is determined based on the two sets of two-dimensional coordinates.

[0114] In this embodiment, for the acquired first and second femoral head images, the computer device can call a femoral head center identifier for processing and output the corresponding coordinates of the femoral head center point. Thus, the two-dimensional images input to the femoral head center identifier include two images, and the corresponding two-dimensional coordinates output by the femoral head center identifier include two sets. Based on these two sets of two-dimensional coordinates, using the viewing angle during image rendering, a point in three-dimensional space can be obtained through projection processing. The coordinates of this point are the femoral head center point in the femoral head image data.

[0115] Reference Figure 5 The diagram illustrates a femoral head center point identification device according to an embodiment of this application, which may specifically include an acquisition module 501, a rendering module 502, a processing module 503, and a determination module 504, wherein:

[0116] The acquisition module 501 is used to acquire the femoral head image data to be identified, wherein the femoral head image data is two-dimensional image data or three-dimensional image data;

[0117] Rendering module 502 is used to render the femoral head image data to obtain a first femoral head image and a second femoral head image, both of which are two-dimensional images.

[0118] Processing module 503 is used to call the femoral head center recognizer to process the first femoral head image and the second femoral head image to obtain two sets of two-dimensional coordinates output by the femoral head center recognizer. The femoral head center recognizer is trained based on multiple two-dimensional femoral head sample images labeled with the femoral head center point.

[0119] The determination module 504 is used to determine the center point of the femoral head in the femoral head image data based on the two sets of two-dimensional coordinates.

[0120] In this embodiment of the application, the rendering module 502 can specifically be used for:

[0121] The femoral head image data is rendered into a three-dimensional space;

[0122] Determine the first-person and second-person perspectives;

[0123] Image acquisition is performed on the femoral head image data in the three-dimensional space based on the first viewpoint and the second viewpoint respectively to obtain the first femoral head image and the second femoral head image.

[0124] In one possible implementation of this application embodiment, the rendering module 502 may also be used for:

[0125] Construct a three-dimensional coordinate system corresponding to the three-dimensional space;

[0126] A target surface in the three-dimensional space is determined, and a first viewpoint is determined in the direction facing the target surface. The target surface is a plane perpendicular to any axis in the three-dimensional coordinate system.

[0127] Determine a second perspective that is orthogonal to the first perspective.

[0128] In another possible implementation of this application embodiment, the rendering module 502 may also be used for:

[0129] Based on the first perspective, the operation of acquiring the femoral head image data using a first camera at the first perspective position is simulated to generate a first femoral head image acquired by the first camera.

[0130] Based on the second perspective, the operation of acquiring the femoral head image data using a second camera at the second perspective position is simulated to generate a second femoral head image acquired by the second camera.

[0131] In this embodiment of the application, the two sets of two-dimensional coordinates include a first coordinate and a second coordinate. The first coordinate is the coordinate of the first femoral head center point output by the femoral head center recognizer for the first femoral head image, and the second coordinate is the coordinate of the second femoral head center point output by the femoral head center recognizer for the second femoral head image. The first viewpoint and the second viewpoint are orthogonal viewpoints. The determining module 504 can specifically be used for:

[0132] Determine a first line from the first viewpoint to the center point of the first femoral head, and a second line from the second viewpoint to the center point of the second femoral head, wherein the extensions of the first line and the second line intersect at a point in space.

[0133] Project the intersection point of the extension of the first line and the extension of the second line onto the three-dimensional space corresponding to the femoral head image data;

[0134] The projection point in the three-dimensional space is identified as the center point of the femoral head in the femoral head image data.

[0135] In one possible implementation of this application embodiment, the determining module 504 may further be used for:

[0136] Determine the spatial matrix and calibration matrix corresponding to the first viewpoint, wherein the first viewpoint is facing the target surface of the femoral head image data, and the target surface is a plane perpendicular to any axis in the coordinate system of the three-dimensional space;

[0137] Based on the spatial matrix and the calibration matrix, matrix calculations are performed on the intersection points to obtain the coordinates of the femoral head center point in the three-dimensional space corresponding to the femoral head image data.

[0138] In this embodiment of the application, the two-dimensional femoral head sample image used to train the femoral head center recognizer includes multiple two-dimensional images obtained by reconstructing multiple femoral head sample image data from multiple perspectives. The number of two-dimensional femoral head sample images reconstructed based on any of the femoral head sample image data is equal to the number of multiple perspectives.

[0139] This application provides a femoral head center point recognition device, which can be a device or a corresponding functional module or unit component within a device capable of performing the corresponding function. For example, the femoral head center point recognition device can be a computer device or a module within a computer device as described in the foregoing embodiments. Using this device, the steps in the foregoing method embodiments can be implemented.

[0140] 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.

[0141] 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 femoral head center point recognition method described above, for example... Figure 1 Steps S101 to S104 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.

[0142] 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 an acquisition module, a rendering module, a processing module, and a determination module, with the specific functions of each module as follows:

[0143] The acquisition module is used to acquire the femoral head image data to be identified, wherein the femoral head image data is two-dimensional image data or three-dimensional image data;

[0144] The rendering module is used to render the femoral head image data to obtain a first femoral head image and a second femoral head image, both of which are two-dimensional images.

[0145] The processing module is used to call the femoral head center recognizer to process the first femoral head image and the second femoral head image to obtain two sets of two-dimensional coordinates output by the femoral head center recognizer. The femoral head center recognizer is trained based on multiple two-dimensional femoral head sample images labeled with the femoral head center point.

[0146] The determination module is used to determine the center point of the femoral head in the femoral head image data based on the two sets of two-dimensional coordinates.

[0147] The computer device 600 may be a device capable of implementing the relevant steps or functions in the foregoing method embodiments. Exemplarily, the computer device 600 may be a desktop computer, a cloud server, or other such 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.

[0148] 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.

[0149] 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.

[0150] 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.

[0151] 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.

[0152] 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.

[0153] 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 identifying the center point of the femoral head, characterized in that, include: Acquire the femoral head image data to be identified, wherein the femoral head image data is two-dimensional image data or three-dimensional image data; The femoral head image data is rendered into a three-dimensional space, and a first viewpoint and a second viewpoint are determined. The femoral head image data in the three-dimensional space is captured based on the first viewpoint and the second viewpoint respectively to obtain a first femoral head image and a second femoral head image. The first femoral head image and the second femoral head image are both two-dimensional images, and the first viewpoint and the second viewpoint are orthogonal viewpoints. The femoral head center identifier is invoked to process the first femoral head image and the second femoral head image to obtain two sets of two-dimensional coordinates output by the femoral head center identifier. The femoral head center identifier is trained based on multiple two-dimensional femoral head sample images labeled with the femoral head center point. The two sets of two-dimensional coordinates include a first coordinate and a second coordinate. The first coordinate is the first femoral head center point coordinate output by the femoral head center identifier for the first femoral head image, and the second coordinate is the second femoral head center point coordinate output by the femoral head center identifier for the second femoral head image. Determine a first line from the first viewpoint to the center point of the first femoral head, and a second line from the second viewpoint to the center point of the second femoral head, wherein the extensions of the first line and the second line intersect at a point in space. Project the intersection point of the extension of the first line and the extension of the second line onto the three-dimensional space corresponding to the femoral head image data; The projection point in the three-dimensional space is identified as the center point of the femoral head in the femoral head image data.

2. The method according to claim 1, characterized in that, Determining the first and second perspectives includes: Construct a three-dimensional coordinate system corresponding to the three-dimensional space; A target surface in the three-dimensional space is determined, and a first viewpoint is determined in the direction facing the target surface. The target surface is a plane perpendicular to any axis in the three-dimensional coordinate system. Determine a second perspective that is orthogonal to the first perspective.

3. The method according to claim 1, characterized in that, The step of acquiring images of the femoral head in the three-dimensional space based on the first and second perspectives respectively, to obtain a first femoral head image and a second femoral head image, includes: Based on the first perspective, the operation of acquiring the femoral head image data using a first camera at the first perspective position is simulated to generate a first femoral head image acquired by the first camera. Based on the second perspective, the operation of acquiring the femoral head image data using a second camera at the second perspective position is simulated to generate a second femoral head image acquired by the second camera.

4. The method according to claim 1, characterized in that, The step of projecting the intersection point of the extensions of the first and second connecting lines onto the three-dimensional space corresponding to the femoral head image data includes: Determine the spatial matrix and calibration matrix corresponding to the first viewpoint; Based on the spatial matrix and the calibration matrix, matrix calculations are performed on the intersection points to obtain the coordinates of the femoral head center point in the three-dimensional space corresponding to the femoral head image data.

5. The method according to any one of claims 1 to 4, characterized in that, The two-dimensional femoral head sample images used to train the femoral head center recognizer include multiple two-dimensional images obtained by reconstructing multiple femoral head sample image data from multiple perspectives. The number of two-dimensional femoral head sample images reconstructed based on any of the femoral head sample image data is equal to the number of multiple perspectives.

6. A femoral head center point identification device, characterized in that, include: The acquisition module is used to acquire the femoral head image data to be identified, wherein the femoral head image data is two-dimensional image data or three-dimensional image data; The rendering module is used to render the femoral head image data into a three-dimensional space, determine a first perspective and a second perspective, and perform image acquisition on the femoral head image data in the three-dimensional space based on the first perspective and the second perspective respectively to obtain a first femoral head image and a second femoral head image. The first femoral head image and the second femoral head image are both two-dimensional images, and the first perspective and the second perspective are orthogonal perspectives. The processing module is used to call the femoral head center recognizer to process the first femoral head image and the second femoral head image to obtain two sets of two-dimensional coordinates output by the femoral head center recognizer. The femoral head center recognizer is trained based on multiple two-dimensional femoral head sample images labeled with the femoral head center point. The two sets of two-dimensional coordinates include a first coordinate and a second coordinate. The first coordinate is the first femoral head center point coordinate output by the femoral head center recognizer for the first femoral head image, and the second coordinate is the second femoral head center point coordinate output by the femoral head center recognizer for the second femoral head image. The determination module is used to determine a first line connecting the first viewpoint to the center point of the first femoral head, and a second line connecting the second viewpoint to the center point of the second femoral head, wherein the extensions of the first and second lines intersect at a point in space; project the intersection point of the extensions of the first and second lines onto the three-dimensional space corresponding to the femoral head image data; and identify the projection point in the three-dimensional space as the center point of the femoral head in the femoral head image data.

7. 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 5.

8. 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 5 is performed.

Images