Cobb angle measurement method and apparatus

Abstract

Provided in the present application are a Cobb angle measurement method and apparatus. The method comprises: receiving from a terminal device an X-ray film to be tested of a target body to be tested; inputting said X-ray film into a first recognition model, so as to obtain a spine recognition region; and on the basis of the spine recognition region, obtaining a Cobb angle of the spine of said target body. Thus, a Cobb angle can be automatically measured, thereby shortening a measurement duration, improving the measurement accuracy, and facilitating accurate analysis of the severity of scoliosis on the basis of the Cobb angle. The apparatus is configured to perform the steps of the method.

Description

Cobb angle measurement method and device

[0001] This application claims priority to the Chinese patent application filed with the State Intellectual Property Office on August 31, 2023, with application number 202311127318.1 and application name “Cobb angle measurement method and device”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of measurement technology, and in particular to a Cobb angle measurement method and device. Background Art

[0003] Scoliosis is a common health condition. Scoliosis severity can be assessed using the curvature angle, one measure of which is the Cobb angle. Therefore, accurate Cobb angle measurement is crucial for assessing scoliosis severity.

[0004] Currently, the traditional Cobb angle measurement method involves observing an X-ray of the user's spine to identify the upper and lower vertebral bodies. The upper and lower vertebral bodies refer to the vertebrae with the greatest inclination toward the concave side of the scoliosis curve. Using a ruler, the examiner draws a horizontal line along the upper edge of the upper vertebral body and a horizontal line along the lower edge of the lower vertebral body, and then draws a perpendicular line between these two lines. Using a protractor, the examiner measures the angle between these two perpendicular lines. This angle is the Cobb angle.

[0005] However, the above method relies on the experience of the tester to determine the upper and lower vertebral bodies, which is prone to subjective judgment errors. The tester also needs to manually draw lines and measure manually. The process is not automatic and prone to errors, resulting in inaccurate measurement of the Cobb angle.

[0006] Summary of the Invention

[0007] The present application provides a Cobb angle measurement method and device, which can automatically extract the spinal contour in the spinal X-ray film, automatically measure the Cobb angle of the spine based on the extracted spinal contour, and realize accurate measurement of the Cobb angle so as to accurately assess the severity of scoliosis.

[0008] In a first aspect, the present application provides a Cobb angle measurement method, the method comprising:

[0009] receiving an X-ray of the target body to be measured from a terminal device, wherein the X-ray includes a spine of the target body to be measured;

[0010] Inputting the X-ray film to be measured into the first recognition model to obtain a spine recognition region, the spine recognition region is used to indicate the outline of the spine of the target body to be measured, and the first recognition model is used to extract the outline of the spine in the X-ray film;

[0011] According to the spinal column identification area, the Cobb angle of the spinal column of the target body to be measured is obtained.

[0012] By the method provided in the first aspect, an X-ray film of a target body to be measured sent by a terminal device can be received, wherein the X-ray film to be measured includes the spine of the target body to be measured, and the Cobb angle of the target body to be measured can be determined. The X-ray film to be measured can be input into a first recognition model using AI / ML technology, and the contour of the spine in the X-ray film to be measured can be identified by the first recognition model to obtain a spine recognition area. The spine recognition area is used to indicate the contour of the spine of the target body to be measured, and the position where the scoliosis of the target body to be measured is most severe can be indicated with the help of the spine recognition area. Based on the spine recognition area, the Cobb angle of the spine of the target body to be measured can be determined. Thus, the Cobb angle of the spine in the X-ray film can be automatically identified. The whole process is convenient, fast and accurate, without relying on the subjective experience of the tester, nor does it require the tester to manually draw lines and manually measure with the help of auxiliary equipment, thus avoiding the problems of subjective judgment errors and inaccurate human measurements, improving the accuracy of measuring the Cobb angle, and helping to accurately analyze the severity of scoliosis based on the Cobb angle.

[0013] In a possible design, the Cobb angle of the spine of the target body to be measured is obtained according to the spine identification area, including:

[0014] Determining a first vector and a second vector based on the spinal column identification area, wherein the direction of the first vector is the direction of the upper edge of the upper vertebral body in the spinal column of the target object to be measured, and the direction of the second vector is the direction of the lower edge of the lower vertebral body in the spinal column of the target object to be measured;

[0015] The angle between the perpendicular line of the first vector and the perpendicular line of the second vector is determined as the Cobb angle.

[0016] Thus, the Cobb angle can be automatically measured by determining the upper and lower vertebral bodies in the spine.

[0017] In one possible design, determining the first vector and the second vector based on the spinal column identification region includes:

[0018] Traversing the spine recognition area to obtain a first pixel point and a second pixel point in each row of pixels, where the first pixel point and the second pixel point are located on the contour of the spine of the target body to be measured;

[0019] Determine the midpoint between the first pixel and the second pixel in each row of pixels;

[0020] Obtaining a first curve according to the plurality of midpoints;

[0021] Determine a tangent vector from the first curve at every preset step length to obtain multiple tangent vectors, where the tangent vector points from the midpoint of the preset step length to the midpoint of the preset step length;

[0022] Determine the angle between each tangent vector and the horizontal line;

[0023] The tangent vector corresponding to the minimum angle is determined as the first vector, and the tangent vector corresponding to the maximum angle is determined as the second vector.

[0024] In one possible design, the method further includes:

[0025] filtering the first curve to obtain a filtered first curve;

[0026] The filtered first curve is determined as the first curve.

[0027] Therefore, the filtered first curve can eliminate noise points on the first curve, thereby enhancing robustness.

[0028] In one possible design, the method further includes:

[0029] Optimize the size and resolution of the X-ray film to be tested to obtain an optimized X-ray film;

[0030] The optimized X-ray film is determined as the X-ray film to be tested.

[0031] In one possible design, the step of determining the first recognition model includes:

[0032] Acquire a sample X-ray film set, the sample X-ray film set including a plurality of sample X-ray films, the sample X-ray films including a spine of a sample target body;

[0033] Obtaining a spine identification region of each sample X-ray film, where the spine identification region is used to indicate the outline of the spine of the sample target body;

[0034] The initial recognition model is trained according to each sample X-ray film and the spinal column recognition area of ​​each sample X-ray film to obtain a first recognition model. The initial recognition model is used to determine the edge of the spinal column in the X-ray film.

[0035] In this way, a first recognition model can be obtained, which can determine the spinal column recognition area in the X-ray film.

[0036] In a second aspect, the present application provides a Cobb angle measurement device, comprising: a module for executing the method of the first aspect and any possible involvement of the first aspect.

[0037] In a third aspect, the present application provides an electronic device, comprising: a processor;

[0038] The processor is used to execute the computer-executable program or instructions in the memory, so that the electronic device executes the method in the first aspect and any possible design of the first aspect.

[0039] In a fourth aspect, the present application provides an electronic device comprising: a memory and a processor; the memory is used to store program instructions; the processor is used to call the program instructions in the memory so that the electronic device executes the method in the first aspect and any possible design of the first aspect.

[0040] In a fifth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor in an electronic device, implements the method in the first aspect and any possible design of the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] FIG1 is a schematic flow chart of a Cobb angle measurement method provided in one embodiment of the present application;

[0042] FIG2 is a schematic flow chart of a Cobb angle measurement method provided in one embodiment of the present application;

[0043] FIG3 is a schematic diagram of a process for determining a first vector and a second vector according to an embodiment of the present application;

[0044] FIG4 is a schematic diagram of a spine contour in a spine identification area provided by an embodiment of the present application;

[0045] FIG5 is a schematic diagram of a spine contour in a spine identification area provided by an embodiment of the present application;

[0046] FIG6 is a schematic diagram of a process for obtaining a first recognition model according to an embodiment of the present application;

[0047] FIG7 is a schematic diagram of a sample X-ray film in a Cobb angle measurement method provided in one embodiment of the present application;

[0048] FIG8 is a schematic structural diagram of a Cobb angle measurement device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0049] In this application, "at least one" refers to one or more, and "plurality" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: the existence of A alone, the existence of A and B at the same time, and the existence of B alone, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a alone, b alone, or c alone can represent: a alone, b alone, c alone, a and b in combination, a and c in combination, b and c in combination, or a, b, and c in combination, where a, b, and c can be single or multiple. In addition, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as indicating or implying relative importance. The directions or positional relationships indicated by terms such as "center", "longitudinal", "lateral", "up", "down", "left", "right", "front", and "back" are based on the directions or positional relationships shown in the accompanying drawings and are only for the convenience of describing the present application and simplifying the description. They do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction. Therefore, they should not be understood as limiting the present application.

[0050] The present application provides a Cobb angle measurement method, device, electronic device, computer-readable storage medium and computer program product. The server uses artificial intelligence (AI) or ML (machine learning) technology to identify the contour of the spine in the X-ray film sent by the terminal device. The server analyzes the contour of the spine based on the identified contour of the spine, and then determines the Cobb angle of the spine. Thus, the Cobb angle can be measured automatically without the need for auxiliary equipment such as rulers and protractors, saving measurement time, avoiding measurement errors caused by insufficient manual experience and tedious manual operations, improving the accuracy of measuring the Cobb angle, and facilitating accurate analysis of the severity of scoliosis based on the Cobb angle.

[0051] Among them, the server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0052] Among them, the terminal device can be a mobile phone, tablet computer, laptop computer, wearable device, vehicle-mounted device, augmented reality (AR) / virtual reality (VR) device, ultra-mobile personal computer (UMPC), netbook, personal digital assistant (PDA), smart TV, smart screen, HDTV, 4K TV, smart speaker, smart projector and other devices. This application does not impose any restrictions on the specific type of terminal device.

[0053] The terminal device and the server can be connected directly or indirectly via wired or wireless communication methods. Wired communication methods include coaxial cables, optical fibers, and digital subscriber lines (DSL). Wireless communication methods include Bluetooth, infrared, Wi-Fi, and microwaves.

[0054] In some embodiments, a network may be used as a medium for communication between a terminal device and a server, and may be a wide area network (WAN) or a local area network (LAN), or a combination thereof. For example, the present application may be implemented with the aid of cloud technology, which refers to a hosting technology that unifies hardware, software, network, and other resources within a WAN or LAN to enable data computing, storage, processing, and sharing.

[0055] Taking the Cobb angle measurement method of the present application being executed by a server as an example, the specific implementation process of the Cobb angle measurement method of the present application by the server will be described in detail below in conjunction with Figure 1.

[0056] Please refer to Figures 1 and 2. Figure 1 shows a schematic flow chart of a Cobb angle measurement method provided in one embodiment of the present application, and Figure 2 shows a schematic flow chart of a Cobb angle measurement method provided in one embodiment of the present application. As shown in Figure 1, the Cobb angle measurement method of the present application may include:

[0057] S101. Receive an X-ray film of a target body to be measured from a terminal device, where the X-ray film includes a spine of the target body to be measured.

[0058] When measuring the Cobb angle of the spine of a subject to be measured, the user can use a terminal device to capture an initial X-ray, so that the terminal device obtains the X-ray of the subject to be measured. The terminal device here has both a capture function and a storage function. Alternatively, the user can store the X-ray of the subject to be measured in the terminal device, which here has a storage function.

[0059] The target object to be tested can be a human, animal, or other organism with a spine. The X-ray film to be tested can be the entire image or a portion of the initial X-ray film. The initial X-ray film refers to the image of the target object obtained after X-ray examination, and the initial X-ray film typically includes all vertebrae in the spine. The X-ray film to be tested includes the spine of the target object to be tested, and the spine here can be all or a portion of the vertebrae in the spine.

[0060] In addition, the resolution of the X-ray film to be measured can be set to be greater than or equal to the preset resolution, which is conducive to the subsequent accurate identification of the Cobb angle of the spine. The preset resolution can be set according to the measurement capabilities of the server, and this application does not limit the specific size of the preset resolution.

[0061] Therefore, after receiving the operation of sending the X-ray film of the target body to be measured, the terminal device can send the X-ray film of the target body to be measured to the server.

[0062] S102. Input the X-ray film to be measured into the first recognition model to obtain a spine recognition area. The spine recognition area is used to indicate the outline of the spine of the target body to be measured. The first recognition model is used to extract the outline of the spine in the X-ray film.

[0063] The first recognition model can identify the outline of the spine in an X-ray. The outline of the spine can represent the overall morphology of the spine. This application does not limit the specific type of the first recognition model. For example, the first recognition model can be an AI or ML model. This application does not limit the specific implementation method of the X-ray, as long as the X-ray includes the spine.

[0064] After receiving the X-ray film to be tested, the server can input the X-ray film to be tested into the first recognition model, and the first recognition model can recognize the outline of the spine in the X-ray film to be tested. In addition, the server can obtain the spine recognition area.

[0065] The spine identification area can indicate the outline of the spine of the subject to be measured. When the X-ray film to be measured includes all vertebrae of the spine, the spine identification area can indicate the outline of all vertebrae, that is, the overall shape of all vertebrae. When the X-ray film to be measured includes only a portion of the vertebrae of the spine, the spine identification area can indicate the outline of the portion of the vertebrae, that is, the overall shape of the portion of the vertebrae.

[0066] The present application does not limit the specific implementation of the spinal column identification region. For example, the spinal column identification region includes at least the region of the spinal column in the X-ray film to be tested, and the spinal column identification region also displays the outline of the spinal column. In addition, the spinal column identification region may also display all or part of the vertebral bodies of the spine.

[0067] In addition, the first recognition model may have requirements on the specifications of the X-ray film to be tested, which is helpful to ensure the accuracy of the recognition results.

[0068] Based on this, the server can adjust the size of the X-ray film to match the size of the first recognition model, such as by enlarging or reducing it, thereby optimizing the size of the X-ray film and improving computing performance. Furthermore, the server can set the reduction or enlargement ratio to 1 by default. And / or, the server can adjust the resolution of the X-ray film to match the resolution of the first recognition model, such as by increasing or decreasing it, thereby optimizing the resolution of the X-ray film.

[0069] In addition to size and resolution, the server can also optimize other parameters of the X-ray film being tested. For example, the server can also perform angle correction on the X-ray film being tested, which helps reduce recognition complexity.

[0070] Thus, the server can obtain the optimized X-ray film and determine the optimized X-ray film as the X-ray film to be tested.

[0071] S103 : Obtain the Cobb angle of the spine of the target body to be measured according to the spine identification area.

[0072] Because the spine's contour represents its overall morphology, the server, using algorithms such as angle functions, can identify the most severe scoliosis location based on spinal region identification. Based on the corresponding scoliosis angle at that location, the server can then determine the Cobb angle of the subject's spine.

[0073] In a specific embodiment, the specific implementation process of the Cobb angle measurement method of the present application is described in detail with reference to FIG2 .

[0074] As shown in FIG2 , the method includes:

[0075] The mobile phone takes a photo of the initial X-ray film of user A to obtain the X-ray film to be tested.

[0076] The mobile phone sends the X-ray film to be tested to the server.

[0077] In response to the requirement of the first recognition model on the size of the X-ray film to be measured, the server detects whether the size of the X-ray film to be measured meets the size required by the first recognition model.

[0078] If not, such as the size of the X-ray film to be tested is large or small, the server can optimize the specification parameters of the X-ray film to be tested, scale the size of the X-ray film to be tested to the size required by the first recognition model, and then the server inputs the scaled X-ray film to be tested into the first recognition model.

[0079] If the conditions are met, the server inputs the X-ray film to be tested into the first recognition model.

[0080] The server processes the X-ray film to be tested or the scaled X-ray film to be tested by using the first recognition model to obtain a spinal column recognition area.

[0081] The server uses algorithms such as angle functions to process the spinal identification area to determine the location where the scoliosis is most severe and the corresponding scoliosis angle at that location.

[0082] The server obtains the Cobb angle of the spine of the target body to be measured according to the scoliosis angle.

[0083] The Cobb angle measurement method provided in this application receives an X-ray of a target object to be measured sent by a terminal device, wherein the X-ray includes the spine of the target object to be measured, and the server can be requested to determine the Cobb angle of the target object to be measured. The server can use AI / ML technology to input the X-ray into a first recognition model, and use the first recognition model to identify the contour of the spine in the X-ray to be measured to obtain a spine identification area. The spine identification area is used to indicate the contour of the spine of the target object to be measured, and the spine identification area can be used to indicate the most severe position of the scoliosis of the target object to be measured. The server can determine the Cobb angle of the spine of the target object to be measured based on the spine identification area. Thus, the Cobb angle of the spine in the X-ray can be automatically identified. The entire process is convenient, fast, and accurate, without relying on the subjective experience of the tester, nor does it require the tester to manually draw lines and measure manually with the help of auxiliary equipment. This avoids the problems of subjective judgment errors and human measurement inaccuracies, improves the accuracy of Cobb angle measurement, and helps to accurately analyze the severity of scoliosis based on the Cobb angle.

[0084] Based on the description of the above embodiment, in S103, the server may adopt various methods to obtain the Cobb angle of the spine of the target body to be measured according to the spine identification area.

[0085] In some examples, obtaining the Cobb angle of the spine of the target body to be measured based on the spine identification region includes:

[0086] Determining a first vector and a second vector based on the spinal column identification area, wherein the direction of the first vector is the direction of the upper edge of the upper vertebral body in the spinal column of the target object to be measured, and the direction of the second vector is the direction of the lower edge of the lower vertebral body in the spinal column of the target object to be measured;

[0087] The angle between the perpendicular line of the first vector and the perpendicular line of the second vector is determined as the Cobb angle.

[0088] Based on the spinal column outline indicated by the spinal column identification area, the server can determine the upper edge of the upper vertebral body and the lower edge of the lower vertebral body. The server determines the direction of the upper edge of the upper vertebral body as the direction of the first vector, and the server determines the direction of the lower edge of the lower vertebral body as the direction of the second vector. Thus, the server can determine the angle between the perpendicular line of the first vector and the perpendicular line of the second vector as the Cobb angle of the spinal column of the target object to be measured.

[0089] In summary, the server can automatically measure the Cobb angle by determining the upper vertebral body and the lower vertebral body in the spine.

[0090] 3 to 5 , the specific implementation process of the server determining the first vector and the second vector according to the spinal column identification area will be described in detail.

[0091] Please refer to Figures 3-5. Figure 3 shows a schematic diagram of the process of determining the first vector and the second vector according to an embodiment of the present application. Figures 4 and 5 show schematic diagrams of a spinal contour in a spinal identification region according to an embodiment of the present application. As shown in Figure 3, determining the first vector and the second vector according to the spinal identification region includes:

[0092] S201 , traverse the spine recognition area to obtain the first pixel point and the second pixel point in each row of pixels, where the first pixel point and the second pixel point are located on the contour of the spine of the target body to be measured.

[0093] The server scans the pixels in the spine identification area row by row to obtain each row of pixels, thereby traversing the spine identification area. Each row of pixels can be represented as a pixel set, which includes all pixels in the row in the spine identification area.

[0094] For each row of pixels in the spine identification area, the server may determine two pixels located on the outline of the spine of the target body to be detected as the first pixel and the second pixel in the row of pixels. The first pixel may be point A shown in FIG4 , and the second pixel may be point B shown in FIG4 .

[0095] S202: Determine the midpoint between the first pixel point and the second pixel point in each row of pixels.

[0096] S203: Obtain a first curve according to the multiple midpoints.

[0097] For the first pixel and the second pixel in each row of pixels, the server may determine the pixel in the row that is equidistant from the first pixel and the second pixel as the midpoint between the first pixel and the second pixel. Thus, the server may obtain multiple midpoints in all rows of pixels.

[0098] For example, in FIG4 , where the first pixel is point A and the second pixel is point B, the midpoint may be point C. Furthermore, the number of midpoints is related to the resolution of the X-ray film being measured. In some examples, the greater the resolution of the X-ray film being measured, the greater the number of midpoints.

[0099] After obtaining multiple midpoints in all rows of pixel points, the server may connect the multiple midpoints in sequence to obtain a first curve.

[0100] S2041: Filter the first curve to obtain a filtered first curve.

[0101] S2042: Determine the filtered first curve as the first curve.

[0102] Among them, S2041 and S2042 are optional steps.

[0103] When obtaining the first curve, considering that the first curve may be relatively smooth, the server may execute S205.

[0104] When obtaining the first curve, considering that the first curve may have some noise points, which may interfere with the measurement result, the server may execute S2041-S2042.

[0105] The server may filter the first curve to obtain a filtered first curve, wherein the filtered first curve may eliminate noise points on the first curve, thereby enhancing robustness.

[0106] In some examples, the server can use a Savitzky-Golay filter to filter the first curve. In selecting the window length (window-length), the following three points need to be considered: First, the curvature angle of the spine of the target body to be measured generally does not appear at a very sharp angle at one point, so the low-frequency component accounts for a large proportion overall. Second, although a window-length that is too long has a better smoothing effect, it may cause the final cobb angle to be distorted. Third, although a window-length that is too short can reduce the distortion of the cobb angle, it may retain too much noise, affecting the accuracy of the final cobb angle. In summary, the window-length can be (sample data length) / 10. If the window-length is an even number, (window-length+1) is used as the final window to ensure that the window-length is an odd number. In practice, the window-length can also be taken from (sample data length) / 10 to (sample data length) / 20.

[0107] Therefore, the server may determine the filtered curve as the first curve.

[0108] S205 , determining a tangent vector from the first curve at every preset step length to obtain multiple tangent vectors, where the tangent vector points from a midpoint at the start of the preset step length to a midpoint at the end of the preset step length.

[0109] The server sequentially obtains midpoints of a preset step length from the first curve to obtain multiple tangent vectors. For any tangent vector, the server can determine the midpoint of the preset step length starting and the midpoint of the preset step length ending from the first curve. In Figure 5, for the first tangent vector D1 among the multiple tangent vectors Dm, C1 is the midpoint of the preset step length starting, C2 is the midpoint of the preset step length ending, and the first tangent vector D1 points from point C1 to point C2. For the second tangent vector D2 among the multiple tangent vectors Dm, C2 is the midpoint of the preset step length starting, C3 is the midpoint of the preset step length ending, and the second tangent vector D2 points from point C2 to point C3. Here, m is a positive integer greater than or equal to 1 and less than or equal to the total number of the multiple tangent vectors.

[0110] This application does not limit the specific size of the preset step size. For example, if the preset step size is 5, then for the first tangent vector, the server can start from the first midpoint of the first curve, obtain five midpoints, and determine the first midpoint and the fifth midpoint of these five midpoints. Thus, the server can determine the vector from the first midpoint to the fifth midpoint as a tangent vector.

[0111] In addition, the preset step size can be set according to the measurement accuracy requirement. In some examples, if the measurement accuracy requirement is higher, the preset step size should be set smaller.

[0112] S206: Determine the angle between each tangent vector and the horizontal line.

[0113] S207 : Determine the tangent vector corresponding to the minimum angle as the first vector, and determine the tangent vector corresponding to the maximum angle as the second vector.

[0114] For any tangent vector, the server can determine the angle between the tangent vector and the horizontal line. Thus, the server can obtain the angles corresponding to all tangent vectors. The angle between a tangent vector and the horizontal line is the angle between the tangent vector and the horizontal line that is less than or equal to 90°.

[0115] Since the upper vertebral body in the spine of the target body to be measured and the lower vertebral body in the spine of the target body to be measured refer to the vertebrae with the largest inclination toward the concave side of scoliosis, and the Cobb angle is the angle between the perpendicular line to the corresponding straight line of the upper edge of the upper vertebral body and the perpendicular line to the horizontal line corresponding to the lower edge of the lower vertebral body, the server can determine the tangent vector corresponding to the minimum angle as the first vector, and the E vector shown in Figure 5 is the tangent vector corresponding to the minimum angle, that is, the first vector. The server can determine the tangent vector corresponding to the maximum angle as the second vector, and the F vector shown in Figure 5 is the tangent vector corresponding to the maximum angle, that is, the second vector.

[0116] Based on the description of the above embodiment, in S102, the server can generate a first recognition model. Of course, other devices can also generate the first recognition model. When the server needs to use the first recognition model, the server can call the first recognition model in the other device to obtain the spinal column identification area. The other device can be a device with strong computing power other than the server, such as a supercomputer.

[0117] 6-7 , the specific implementation process of the server generating the first recognition model is described in detail.

[0118] Please refer to Figures 6 and 7. Figure 6 shows a schematic diagram of a process for obtaining a first recognition model according to an embodiment of the present application, and Figure 7 shows a schematic diagram of a sample X-ray film in a Cobb angle measurement method according to an embodiment of the present application. As shown in Figure 6, the method for obtaining the first recognition model according to the present application may include:

[0119] S301. Acquire a sample X-ray film set, where the sample X-ray film set includes multiple sample X-ray films, and the sample X-ray films include the spine of a sample target body.

[0120] S302: Acquire a spine identification region of each sample X-ray film, where the spine identification region is used to indicate the outline of the spine of the sample target body.

[0121] This application does not limit the number of sample X-ray films in the sample X-ray film set. In some examples, the number of sample X-ray films can be as large as necessary to ensure a thorough training process and accurate training results. Furthermore, any parameters such as size, resolution, angle, and sample target volume of the sample X-ray films in the sample X-ray film set can be the same or different, and this application does not impose any restrictions on this.

[0122] A large number of sample X-ray films may be stored locally on the server. Based on this, the server can obtain the sample X-ray film set from the server's local storage. Alternatively, a large number of sample X-ray films may be stored on other devices. Based on this, the server can initiate a request to the other device to obtain the sample X-ray film set. The other device may be a device with storage capabilities.

[0123] In addition to the sample X-ray film set, the server may also obtain a spinal identification region from each sample X-ray film. This application does not limit the specific implementation of how the spinal identification region indicates the outline of the spine of the sample target body. In some examples, the spinal identification region from any sample X-ray film may be region A shown in FIG7 .

[0124] In addition, for any spinal identification region in a sample X-ray film, the spinal identification region can be displayed and stored in the corresponding sample X-ray film, can be stored together with the corresponding sample X-ray film, or can be stored separately from the corresponding sample X-ray film, and this application does not limit this. It should be noted that there is no temporal order between the above S301 and S302, and S301 and S302 can be executed simultaneously or sequentially.

[0125] S303 : Training an initial recognition model according to each sample X-ray film and the spine recognition region of each sample X-ray film to obtain a first recognition model. The initial recognition model is used to determine the edge of the spine in the X-ray film.

[0126] The initial recognition model has a semantic segmentation function, which can distinguish pixels or regions with different semantics. This application does not limit the specific implementation of the initial recognition model. For example, the initial recognition model can be a PointRend model.

[0127] Based on this, the server can train the initial recognition model based on each sample X-ray film and the spinal recognition region of each sample X-ray film, so that the initial recognition model can learn how to identify the spinal recognition region in the X-ray film. In this way, the server can obtain a first recognition model that can determine the spinal recognition region in the X-ray film.

[0128] In summary, the server can generate a first recognition model.

[0129] Illustratively, the present application provides a Cobb angle measurement device.

[0130] Please refer to Figure 8, which shows a schematic structural diagram of a Cobb angle measurement device provided by an embodiment of the present application. As shown in Figure 8, the Cobb angle measurement device of the present application may include:

[0131] The receiving module 801 is configured to receive an X-ray of the target object to be measured from a terminal device when measuring the Cobb angle, wherein the X-ray includes the spine of the target object to be measured;

[0132] Recognition module 802, configured to input the X-ray film to be tested into a first recognition model to obtain a spinal column recognition region, wherein the spinal column recognition region is used to indicate the outline of the spinal column of the subject to be tested. The first recognition model is used to extract the outline of the spinal column from the X-ray film;

[0133] The determination module 803 is used to obtain the spinal Cobb angle of the target body to be measured according to the spinal identification area.

[0134] In some examples, the Cobb angle measurement device, based on the structure shown in FIG8 , may further include: an optimization module.

[0135] The optimization module is configured to filter the first curve to obtain a filtered first curve; and determine the filtered first curve as the first curve.

[0136] The optimization module is also used to optimize the size and resolution of the X-ray film to be tested to obtain an optimized X-ray film; and determine the optimized X-ray film as the X-ray film to be tested.

[0137] In some examples, the determination module 803 is specifically used to determine a first vector and a second vector based on a spinal identification area, wherein the direction of the first vector is the direction of the upper edge of the upper vertebral body in the spine of the target body to be measured, and the direction of the second vector is the direction of the lower edge of the lower vertebral body in the spine of the target body to be measured; and the angle between the perpendicular line of the first vector and the perpendicular line of the second vector is determined as the Cobb angle.

[0138] In some examples, the determination module 803 is specifically used to traverse the spine identification area, obtain the first pixel point and the second pixel point in each row of pixels, and the first pixel point and the second pixel point are located on the contour of the spine of the target body to be measured; determine the midpoint of the first pixel point and the second pixel point in each row of pixels; obtain a first curve based on multiple midpoints; determine a tangent vector from the first curve at every preset step size to obtain multiple tangent vectors, and the tangent vector points from the midpoint at the beginning of the preset step size to the midpoint at the end of the preset step size; determine the angle between each tangent vector and the horizontal line; determine the tangent vector corresponding to the minimum angle as the first vector, and determine the tangent vector corresponding to the maximum angle as the second vector.

[0139] In some examples, the recognition module 802 is further configured to determine a first recognition model, where the determining step includes:

[0140] A sample X-ray film set is obtained, where the sample X-ray film set includes a plurality of sample X-ray films, and the sample X-ray films include a spine of a sample target body.

[0141] A spine identification region of each sample X-ray film is obtained, where the spine identification region is used to indicate the outline of the spine of the sample target body.

[0142] The initial recognition model is trained according to each sample X-ray film and the spinal column recognition area of ​​each sample X-ray film to obtain a first recognition model. The initial recognition model is used to determine the edge of the spinal column in the X-ray film.

[0143] Illustratively, the present application provides an electronic device, comprising: a processor; the processor is configured to execute a computer-executable program or instruction in a memory, so that the electronic device executes the method in the foregoing embodiment.

[0144] Illustratively, the present application provides an electronic device, comprising: a memory and a processor; the memory is used to store program instructions; the processor is used to call the program instructions in the memory so that the electronic device executes the method in the above embodiment.

[0145] Illustratively, the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor in an electronic device, the method in the foregoing embodiment is implemented.

[0146] Exemplarily, the present application provides a computer program product, comprising: execution instructions, the execution instructions are stored in a readable storage medium, at least one processor of an electronic device can read the execution instructions from the readable storage medium, and at least one processor executes the execution instructions so that the electronic device implements the method in the foregoing embodiment.

[0147] In the above embodiments, all or part of the functions can be implemented by software, hardware, or a combination of software and hardware. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function according to the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrations. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state drive (SSD)).

[0148] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by a computer program instructing related hardware to perform the processes. The program can be stored in a computer-readable storage medium, and when executed, the program can include the processes in the above-described method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0149] Finally, it should be noted that the above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A Cobb angle measurement method, characterized in that: The method comprises: Receiving an X-ray film of the target body to be measured from a terminal device, wherein the X-ray film to be measured includes a spine of the target body to be measured; Inputting the X-ray film to be tested into a first recognition model to obtain a spine recognition region, wherein the spine recognition region is used to indicate the contour of the spine of the target body to be tested, and the first recognition model is used to extract the contour of the spine in the X-ray film; The Cobb angle of the spine of the target body to be measured is obtained according to the spine identification area.

2. The method according to claim 1, characterized in that The step of obtaining the Cobb angle of the spine of the target body to be measured according to the spine identification area includes: Determine a first vector and a second vector according to the spine identification area, wherein the direction of the first vector is the direction of the upper edge of the upper vertebral body in the spine of the target body to be measured, and the direction of the second vector is the direction of the lower edge of the lower vertebral body in the spine of the target body to be measured; The angle between the perpendicular line of the first vector and the perpendicular line of the second vector is determined as the Cobb angle.

3. The method according to claim 2, characterized in that The step of determining the first vector and the second vector according to the spinal column identification area comprises: Traversing the spine identification area, obtaining a first pixel point and a second pixel point in each row of pixel points, wherein the first pixel point and the second pixel point are located on the contour of the spine of the target body to be detected; Determine a midpoint between the first pixel point and the second pixel point in each row of pixel points; Obtaining a first curve according to the plurality of midpoints; Determine a tangent vector from the first curve at every preset step length to obtain a plurality of tangent vectors, wherein the tangent vector points from a midpoint at the beginning of the preset step length to a midpoint at the end of the preset step length; Determine the angle between each tangent vector and the horizontal line; The tangent vector corresponding to the minimum angle is determined as the first vector, and the tangent vector corresponding to the maximum angle is determined as the second vector.

4. The method according to claim 3, characterized in that The method further comprises: filtering the first curve to obtain a filtered first curve; The filtered first curve is determined as the first curve.

5. The method according to claim 1, characterized in that The method further comprises: Optimizing the size and resolution of the X-ray film to be tested to obtain an optimized X-ray film; The optimized X-ray film is determined as the X-ray film to be tested.

6. The method according to claim 1, characterized in that The step of determining the first recognition model comprises: Acquire a sample X-ray film set, wherein the sample X-ray film set includes a plurality of sample X-ray films, and the sample X-ray films include a spine of a sample target body; Acquire a spine identification region of each sample X-ray film, wherein the spine identification region is used to indicate the outline of the spine of the sample target body; The initial recognition model is trained according to each sample X-ray film and the spine recognition area of ​​each sample X-ray film to obtain the first recognition model, and the initial recognition model is used to determine the edge of the spine in the X-ray film.

7. A Cobb angle measuring device, characterized in that: The device comprises: a module for executing the Cobb angle measurement method according to any one of claims 1-6.

8. An electronic device, characterized in that: include: processor; The processor is used to execute the computer executable program or instructions in the memory, so that the electronic device executes the Cobb angle measurement method according to any one of claims 1 to 6.

9. An electronic device, characterized in that: include: at least one memory and at least one processor; The memory is used to store computer executable programs or instructions; The processor is used to call the computer executable program or instruction in the memory, so that the electronic device executes the Cobb angle measurement method according to any one of claims 1-6.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer-executable program or instruction, and the computer-executable program or instruction is configured to execute the Cobb angle measurement method according to any one of claims 1 to 6.

Images