Auxiliary diagnosis and evaluation system for teenager idiopathic scoliosis based on X-ray film

Abstract

The invention discloses a teenager idiopathic scoliosis auxiliary diagnosis and evaluation system based on an X-ray film, and relates to the technical field of medical instruments. Comprising an image preprocessing module, a key point detection module, a Cobb angle measurement module and a Lenke typing module. According to the teenager idiopathic scoliosis auxiliary diagnosis and evaluation system based on the X-ray film, key point detection, automatic Cobb angle measurement and Lenke typing including bending type, lumbar vertebra correction type and thoracic vertebra sagittal view kyphosis typing of the standing position, the Bending position and the lateral position spine X-ray film are achieved, the accuracy, efficiency and reliability of auxiliary diagnosis are remarkably improved, and the system is suitable for popularization and application. And comprehensive support is provided for clinical diagnosis and treatment decision of scoliosis.

Description

Technical Field

[0001] The present invention belongs to the technical field of medical devices, and in particular relates to an auxiliary diagnosis and evaluation system for adolescent idiopathic scoliosis based on X-rays. Background Art

[0002] Scoliosis is a common three-dimensional spinal deformity that refers to a lateral curvature of the spine along its longitudinal axis. It is often characterized by lateral deviation of the spine and loss of normal sagittal curvature. The gold standard for diagnosis is a Cobb angle greater than 10°. Normally, the human spine is a straight line, but when scoliosis occurs, it causes the spine to assume a "C" or "S" shape on the plane. The disease typically develops during childhood and adolescence, with adolescent idiopathic scoliosis (AIS) being the most common, accounting for over 80% of scoliosis cases. AIS has an insidious onset, and if not detected and treated promptly, it can lead to problems such as body asymmetry, abnormal posture, and back pain. In more severe cases, it can even affect cardiopulmonary function. Therefore, early detection, diagnosis, and intervention can effectively prevent the progression of scoliosis and are crucial for maintaining good health.

[0003] Current scoliosis diagnosis primarily involves physical examination, X-rays, and other imaging tests. X-rays are currently the most commonly used imaging modality for scoliosis assessment in clinical medicine. By taking spinal X-rays in the anteroposterior and lateral bending positions, the Cobb angle (the angle between the starting and ending vertebrae) can be measured. Originally proposed by American physician John Robert Cobb in 1948 to assess scoliosis severity, the Cobb angle is a key indicator of scoliosis severity and forms the basis of the Lenke classification.

[0004] The existing Cobb angle measurement technologies mainly include: 1) Manual measurement: The traditional Cobb angle measurement method is that the doctor manually selects the vertebrae in the spine that are most severely tilted toward the concave side of scoliosis as the upper and lower end vertebrae, and then uses a protractor to measure the angle between the upper and lower end vertebrae of the spine. 2) Semi-automatic measurement of Cobb angle: Zhang proposed a semi-automatic computer-assisted measurement algorithm for Cobb angle based on neural network technology. After manually selecting the area of ​​interest blocks where the upper and lower end vertebrae are located, the corresponding slope of the spine is automatically estimated to calculate the Cobb angle. 3) Fully automatic measurement of Cobb angle: Tu proposed an automatic Cobb angle measurement algorithm based on traditional machine learning for X-ray films. This method uses a segmentation algorithm to segment the spine and extract the center points of each vertebra. Then, a sixth-order polynomial is used to fit the set of center points to obtain the spinal curve, and then the sixth-order polynomial is used to automatically calculate the Cobb angle.

[0005] While existing Cobb angle measurement techniques have aided scoliosis diagnosis to some extent, they still have significant drawbacks. Manual Cobb angle measurement relies on the physician's manual selection of the starting and ending vertebrae, which is susceptible to personal experience and drawing habits and is subject to significant subjectivity. Semi-automated Cobb angle measurement still requires the physician's involvement in key steps to verify and adjust the results, and cannot completely eliminate subjective influences. Fully automated Cobb angle measurement, typically based on image processing, machine learning, and deep learning technologies, enables automatic identification, segmentation, and measurement of spinal curves in images. However, this approach also presents several drawbacks and challenges: 1) Segmentation Accuracy: Segmentation of the spinal bones is a key step in fully automated measurement methods. However, segmentation accuracy can be affected by factors such as morphological and density variations of the spinal bones in images, noise interference, and occlusion between soft tissue and bone. Particularly in complex cases or low-contrast images, segmentation algorithms may struggle to accurately identify and separate the spinal bones, leading to inaccurate measurement results. 2) Inaccurate Keypoint Detection: Due to image quality, contrast, and anatomical complexity, existing technologies lack accuracy in automatically detecting vertebral keypoints. 3) Poor adaptability to different posture radiographs: Most existing technologies are only applicable to standing radiographs, with limited analysis capabilities for bending and lateral radiographs, restricting their application in different clinical scenarios. 4) Lack of automated Lenke classification: The lack of automated Lenke classification tools requires physicians to manually classify based on measurement results, increasing the complexity and time-consuming nature of the diagnostic process. Summary of the Invention

[0006] In order to solve the above problems existing in the prior art, the present invention provides an auxiliary diagnosis and evaluation system for adolescent idiopathic scoliosis based on X-rays. The technical problem to be solved by the present invention is achieved through the following technical solutions:

[0007] The present invention provides an X-ray-based auxiliary diagnosis and evaluation system for adolescent idiopathic scoliosis, comprising:

[0008] An image preprocessing module is used to perform ROI detection on the spinal X-ray film to be detected, obtain a cut-off image containing the spine, and perform size adjustment and normalization processing on the cut-off image to obtain a preprocessed image;

[0009] A key point detection module, configured to detect the pre-processed image using a preset trained key point detection model to obtain the coordinates of the key points of each vertebra in the pre-processed image;

[0010] A Cobb angle measurement module is used to calculate the standing Cobb angle, the bending Cobb angle and the lateral angle according to the coordinates of the key points of each vertebra in the preprocessed image;

[0011] The Lenke classification module is used to perform Lenke classification based on the spinal X-ray film to be tested, the standing Cobb angle, the bending Cobb angle and the lateral angle to obtain a Lenke classification result.

[0012] Compared with the prior art, the present invention has the following beneficial effects:

[0013] 1. The X-ray-based auxiliary diagnosis and evaluation system for adolescent idiopathic scoliosis of the present invention realizes the key point detection of standing, bending and lateral spinal X-rays, automated Cobb angle measurement and Lenke classification, including curve type, lumbar modified type and thoracic sagittal kyphosis classification, significantly improving the accuracy, efficiency and reliability of auxiliary diagnosis, and providing comprehensive support for the clinical diagnosis and treatment decision-making of scoliosis.

[0014] 2. The X-ray-based adolescent idiopathic scoliosis auxiliary diagnosis and evaluation system of the present invention, through deep learning and key point detection technology, can accurately identify the key points of the vertebrae in various postures, including standing, left and right bending, and lateral positions, thereby calculating a more accurate Cobb angle and significantly improving the accuracy of the Cobb angle measurement results.

[0015] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the following preferred embodiments are specifically cited and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a structural block diagram of an X-ray-based auxiliary diagnosis and evaluation system for adolescent idiopathic scoliosis provided by an embodiment of the present invention;

[0017] Figure 2 This is a processing flow chart of an image preprocessing module provided by an embodiment of the present invention;

[0018] Figure 3 This is a processing flow chart of a key point detection module provided by an embodiment of the present invention;

[0019] Figure 4 Schematic diagram of a key point detection model provided by an embodiment of the present invention;

[0020] Figure 5 This is a flow chart for calculating the Cobb angle in a standing position provided by an embodiment of the present invention;

[0021] Figure 6This is a calculation flow chart of the Cobb angle of the bending position provided by an embodiment of the present invention;

[0022] Figure 7 This is a bending type parting flow chart provided by an embodiment of the present invention;

[0023] Figure 8 This is a classification flow chart of a lumbar vertebra correction type provided by an embodiment of the present invention;

[0024] Figure 9 This is an example diagram of the relationship between a CSVL and a pedicle provided by an embodiment of the present invention;

[0025] Figure 10 It is a schematic diagram of the classification of the lumbar vertebra correction type provided by an embodiment of the present invention.

[0026] Figure 11 This is a flow chart for classifying thoracic sagittal kyphosis provided by an embodiment of the present invention;

[0027] Figure 12 This is a structural block diagram of another X-ray-based auxiliary diagnosis and evaluation system for adolescent idiopathic scoliosis provided by an embodiment of the present invention;

[0028] Figure 13 This is a diagram of a Cobb angle measurement interface for a standing position provided by an embodiment of the present invention;

[0029] Figure 14 This is a diagram of a measurement interface for a bending position Cobb angle provided by an embodiment of the present invention;

[0030] Figure 15 This is an interface diagram of a lateral angle and Lenke classification provided by an embodiment of the present invention;

[0031] Figure 16 This is an interactive flow chart of visualization software provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0032] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the following is a detailed description of an X-ray-based auxiliary diagnosis and evaluation system for adolescent idiopathic scoliosis proposed in accordance with the present invention, in combination with the accompanying drawings and specific implementation methods.

[0033] The aforementioned and other technical contents, features, and effects of the present invention are clearly presented in the following detailed description of the specific embodiments in conjunction with the accompanying drawings. Through the description of the specific embodiments, a deeper and more specific understanding of the technical means and effects adopted by the present invention to achieve the intended purpose can be obtained. However, the accompanying drawings are provided for reference and illustration purposes only and are not intended to limit the technical solutions of the present invention.

[0034] The embodiment of the present invention provides an auxiliary diagnosis and evaluation system for adolescent idiopathic scoliosis based on X-rays, see Figure 1 , Figure 1 This is a structural block diagram of an X-ray-based auxiliary diagnosis and evaluation system for adolescent idiopathic scoliosis provided by an embodiment of the present invention. Figure 1 As shown, the X-ray-based adolescent idiopathic scoliosis auxiliary diagnosis and assessment system of this embodiment includes: an image preprocessing module, a key point detection module, a Cobb angle measurement module, and a Lenke classification module. The X-ray-based adolescent idiopathic scoliosis auxiliary diagnosis and assessment system of this embodiment of the present invention detects key points of the vertebrae on the spinal X-ray to be tested, automatically measures the coronal Cobb angle, and simultaneously calculates the corresponding kyphosis angle on the lateral image. Based on the Cobb and kyphosis angles, the structural nature of the primary and secondary curves is determined, thereby performing the Lenke classification of the scoliosis.

[0035] Furthermore, each module of the adolescent idiopathic scoliosis auxiliary diagnosis and evaluation system based on X-rays in this embodiment is described in detail.

[0036] The image preprocessing module is used to perform ROI (Region of Interest) detection on the spinal X-ray film to be detected, obtain a cut-off image containing the spine, and perform size adjustment and normalization processing on the cut-off image to obtain a preprocessed image.

[0037] In this embodiment, the spinal X-ray films to be detected include: a standing spinal X-ray film, a left bending spinal X-ray film, a right bending spinal X-ray film, and a lateral spinal X-ray film.

[0038] See Figure 2 , Figure 2 is a processing flow chart of an image preprocessing module provided by an embodiment of the present invention, such as Figure 2 As shown in the figure, first, a spinal X-ray film to be detected is selected, and a ROI area detection is performed on it. The intercepted image containing the spine is cut out, and then the intercepted image is adjusted to the input size specified by the key point detection model in the subsequent key point detection module, such as 640×640. After that, normalization processing is performed, that is, the pixel value of the intercepted image is scaled from [0,255] to [0,1] to maintain the same statistical characteristics as the training data of the key point detection model, which is conducive to the stable convergence of the model and improved performance.

[0039] The image preprocessing module of this embodiment uses ROI region detection technology to identify and crop the spinal region in the X-ray film, which can reduce interference information and improve the accuracy of subsequent processing.

[0040] The key point detection module is used to detect the pre-processed image using a preset trained key point detection model to obtain the coordinates of the key points of each vertebra in the pre-processed image.

[0041] In this embodiment, the keypoint detection model is a heatmap regression-based keypoint detection model that introduces a channel attention mechanism. Keypoint heatmap technology is used to obtain the coordinates of the keypoints of each vertebra in the preprocessed image. The keypoints include the upper left, upper right, lower left, and lower right points of the vertebra.

[0042] See Figure 4 , Figure 4 is a structural diagram of a key point detection model provided by an embodiment of the present invention, such as Figure 4 As shown in the figure, the key point detection model uses ResNet34 as the backbone of the network to extract high-level semantic features, and combines deep and shallow features through skip connections to utilize the high-level semantic information and low-level detail information of the image. By constructing a heat map, center offset and angle offset to locate the four corner landmarks of each vertebra, the key points of all vertebrae are finally detected, that is, 17×4 key points from top to bottom vertebrae T1-T12 and L1-L5. In this embodiment, an attention mechanism is added after the deep and shallow features are connected through skip connections ( Figure 4 The yellow part in represents the feature map optimized after the introduction of the channel attention mechanism. The schematic diagram of the channel attention mechanism is as follows Figure 4 The purpose is to make the model focus more on the spinal region to be detected. The improved model has better recognition ability of the vertebra itself and the relationship between the upper and lower vertebrae.

[0043] In this embodiment, the key point detection model is trained using a spinal X-ray training set with 17×4 key points annotated by doctors, where the spinal X-ray training set includes: standing spinal X-rays, left bending spinal X-rays, right bending spinal X-rays, and lateral spinal X-rays.

[0044] See Figure 3 , Figure 3 This is a processing flow chart of a key point detection module provided by an embodiment of the present invention, such as Figure 3 As shown in the figure, taking the standing spinal X-ray as an example, the preprocessed image of the standing spinal X-ray is input into the trained key point detection model for key point detection. First, 17 vertebral center points are detected based on the heat map regression method, that is, these 17 points correspond to 17 vertebrae respectively, completing the positioning of the vertebrae, and then the vertebral center points and the learned angular offset are used to locate the 4 key points around them. The purpose of this is to make the detected vertebral key points more accurate.

[0045] The key point detection module of this embodiment improves the key point detection model by introducing a channel attention mechanism to achieve key point detection of spinal X-rays. The introduced channel attention mechanism enhances the model's response to image features, especially the key point detection performance in low-contrast or blurred images.

[0046] It is understood that in addition to the key point detection technology mentioned in this embodiment, other automated measurement methods can also be used. For example, deep learning segmentation networks, such as U-Net or its variants, can be used to segment the vertebral edges to extract accurate vertebral contours, thereby enabling subsequent automatic measurement of the Cobb angle. Alternatively, pattern recognition algorithms can be applied to identify and classify different scoliosis types to assist in subsequent Cobb angle measurement. Alternatively, multi-task learning can be used to simultaneously predict vertebral segmentation and key points, improving the efficiency and accuracy of the overall subsequent Cobb angle measurement process.

[0047] Among them, the Cobb angle measurement module is used to calculate the standing Cobb angle, bending Cobb angle and lateral angle according to the coordinates of the key points of each vertebra in the preprocessed image.

[0048] In an optional embodiment, the Cobb angle measurement module includes: a standing Cobb angle measurement unit, a bending Cobb angle measurement unit and a lateral angle measurement unit.

[0049] In this embodiment, the standing Cobb angle measurement unit is used to determine the first inclination vector of each vertebra based on the coordinates of the key points of each vertebra in the preprocessed image corresponding to the standing spinal X-ray film, and calculate the angle between the two vertebrae with the largest angle among all vertebrae based on the first inclination vector as the standing Cobb1 angle. The standing Cobb2 angle and the standing Cobb3 angle are calculated based on the scoliosis type of the spinal X-ray film to be detected and the position of the vertebra corresponding to the standing Cobb1 angle.

[0050] Specifically, the key point detection module can be used to obtain the coordinates of the key points of each vertebra in the preprocessed image corresponding to the standing spinal X-ray film, a total of 17×4 key point coordinates. Through the coordinates of the upper left point, lower left point, upper right point and lower right point of each vertebra, the two point coordinates of the left midpoint and the right midpoint of each vertebra can be found. Then the coordinates of the left midpoint are subtracted from the coordinates of the right midpoint to obtain a vector, namely the first tilt vector, the direction of which represents the tilt direction of the vertebra. After obtaining the 17 first tilt vectors corresponding to the 17 vertebrae, the 17×17 matrix can be used to calculate the angle between the two vertebrae, and the two vertebrae with the largest angle and their angle are found, which is the standing Cobb1 angle. Among them, the angle between the two vertebrae is obtained by the dot product of the corresponding vectors of the two vertebrae.

[0051] In this embodiment, the following steps may be included to calculate the standing Cobb2 angle and the standing Cobb3 angle based on the scoliosis type in the spinal X-ray to be detected and the vertebral position corresponding to the standing Cobb1 angle:

[0052] Step 1: Determine the scoliosis type of the spinal X-ray to be detected based on the coordinates of the key points of each vertebra in the preprocessed image corresponding to the spinal X-ray in the standing position.

[0053] Scoliosis types include C and S. Vector calculations can be used to determine the degree of left-right spinal deviation and whether it can be offset, thereby determining the scoliosis type. The specific determination process is as follows: Based on the coordinates of the key points of the 17 vertebrae, the coordinates of the left and right midpoints of each vertebra are obtained. For each point, the difference between its vertical position and the last point is calculated, normalized to the difference between the first and last points. For each point, the difference between its horizontal position and the last point is calculated, normalized to the difference between the first and last points. This yields the normalized coordinate differences for each point on the X and Y axes. For each point, the normalized Y-axis difference is subtracted from the X-axis difference, resulting in a list of differences for all points. Let a be the sum of the difference correlation matrices for all points, and b be the sum of their absolute values. If the difference between a and b is close to 0 (less than 1e-4), the scoliosis type is considered C; otherwise, it is S.

[0054] The spinal morphology is determined by comparing the horizontal and vertical differences between the midpoints of each cone. If the spinal curvature is relatively stable in one direction, known as a C-shape, the difference between a and b will be small. Conversely, if the curvature is complex in two directions, known as an S-shape, the difference will be significantly larger.

[0055] Step 2: If the scoliosis type is type C, the angle between the upper vertebra and vertebra T1 obtained from the standing Cobb1 angle is calculated based on the first tilt vector as the standing Cobb2 angle, and the angle between the lower vertebra and vertebra L5 obtained from the standing Cobb1 angle is calculated as the standing Cobb3 angle.

[0056] Step 3: If the scoliosis type is S-type, determine whether the Cobb1 angle in the standing position is an upper thoracic curve.

[0057] In this embodiment, the curvature type is determined based on the location of the apical vertebra. The apical vertebra at vertebrae T2-T5 is an upper thoracic curve, the apical vertebra at vertebrae T5-T11 is a primary thoracic curve, the apical vertebra at vertebrae T12 is a thoracolumbar curve, and the apical vertebra at vertebrae L1-L4 is a lumbar curve. The apical vertebra is the vertebra farthest from the vertical line of the central sacrum within the scoliosis. Therefore, to determine whether the standing Cobb1 angle is an upper thoracic curve, the apical vertebra of the upper thoracic curve is located and then determined to be within the two vertebrae of the standing Cobb1 angle.

[0058] Specifically, the top vertebra is found according to the following steps: First, the coordinates of the left and right midpoints of the 17 vertebrae are fitted to a fifth-order polynomial curve, representing the overall shape of the spine. The derivative of the fitted polynomial is then derived, and the zero point of the derivative is the extreme point of the curve. Secondly, by finding the zero point of the derivative, the extreme points of the spinal curve are determined, indicating that the shape of the spine has changed at these points. Finally, the range of the extreme points is restricted to ensure that the analysis focuses on the midsection of the spine and eliminates interference from the boundary areas.

[0059] Step 4: If the standing Cobb1 angle is an upper thoracic curve, then based on the first tilt vector, calculate the maximum angle between the lower vertebra of the standing Cobb1 angle and all the vertebrae below the lower vertebra of the standing Cobb1 angle as the standing Cobb2 angle, and calculate the maximum angle between the lower vertebra of the standing Cobb2 angle and all the vertebrae below the lower vertebra of the standing Cobb2 angle as the standing Cobb3 angle.

[0060] Step 5: If the standing Cobb1 angle is not the upper thoracic curve, then based on the first inclination vector, calculate the maximum angle between the upper vertebra of the standing Cobb1 angle and all the vertebrae above the upper vertebra of the standing Cobb1 angle as the standing Cobb2 angle, and calculate the maximum angle between the lower vertebra of the standing Cobb1 angle and all the vertebrae below the lower vertebra of the standing Cobb1 angle as the standing Cobb3 angle.

[0061] For example, the calculation process of the standing Cobb angle is as follows: Figure 5 As shown, Figure 5 This is a flow chart for calculating the Cobb angle in a standing position provided by an embodiment of the present invention.

[0062] In this embodiment, the bending position Cobb angle measurement unit is used to calculate the bending position Cobb1 angle, the bending position Cobb2 angle and the bending position Cobb3 angle based on the coordinates of the key points of each vertebra in the preprocessed image corresponding to the left bending position spinal X-ray, the coordinates of the key points of each vertebra in the preprocessed image corresponding to the right bending position spinal X-ray, and the vertebral positions corresponding to the standing position Cobb1 angle, the standing position Cobb2 angle and the standing position Cobb3 angle.

[0063] Specifically, the following steps may be included:

[0064] Step 1: Determine the third tilt vector of each vertebra based on the coordinates of the key points of each vertebra in the preprocessed image corresponding to the left bending spinal X-ray. Calculate the left bending Cobb1 angle, left bending Cobb2 angle, and left bending Cobb3 angle using the third tilt vector based on the vertebral positions corresponding to the standing Cobb1 angle, standing Cobb2 angle, and standing Cobb3 angle.

[0065] Step 2: Determine the fourth tilt vector of each vertebra based on the coordinates of the key points of each vertebra in the preprocessed image corresponding to the right bending spinal X-ray film. Use the fourth tilt vector to calculate the right bending Cobb1 angle, right bending Cobb2 angle, and right bending Cobb3 angle based on the vertebral positions corresponding to the standing Cobb1 angle, standing Cobb2 angle, and standing Cobb3 angle.

[0066] Step 3: Select the smaller value between the left bending Cobb1 angle and the right bending Cobb1 angle as the bending Cobb1 angle, select the smaller value between the left bending Cobb2 angle and the right bending Cobb2 angle as the bending Cobb2 angle, and select the smaller value between the left bending Cobb3 angle and the right bending Cobb3 angle as the bending Cobb3 angle.

[0067] In this embodiment, the calculation process of the third tilt vector and the fourth tilt vector is similar to the calculation process of the first tilt vector, and will not be described in detail here. For example, taking the calculation of the Cobb angle of the left bending position as an example, the calculation process is as follows: Figure 6 As shown, Figure 6 This is a calculation flow chart of the bending position Cobb angle provided by an embodiment of the present invention.

[0068] In this embodiment, the lateral angle measurement unit is used to determine the second inclination vector of each vertebra based on the coordinates of the key points of each vertebra in the preprocessed image corresponding to the lateral spinal X-ray film, and calculate the angle between vertebra T2 and vertebra T5 as the first lateral angle, the angle between vertebra T10 and vertebra L2 as the second lateral angle, and the angle between vertebra T5 and vertebra T12 as the third lateral angle based on the second inclination vector.

[0069] Among them, the Lenke classification module is used to perform Lenke classification based on the spinal X-ray film to be tested, the standing Cobb angle, the bending Cobb angle and the lateral angle to obtain the Lenke classification result.

[0070] In an optional embodiment, the Lenke classification module includes: a first classification unit, a second classification unit, and a third classification unit. The first classification unit is used to implement the classification of the curvature type, the second classification unit is used to implement the classification of the lumbar modified type, and the third classification unit is used to implement the classification of the thoracic sagittal kyphosis.

[0071] For example, the bending type classification flow chart is as follows: Figure 7 As shown, Figure 7 This is a flow chart for determining a curvature type provided by an embodiment of the present invention. In this embodiment, a first classification unit is configured to determine whether the Cobb angle of the main thoracic curve in the standing position is the largest based on the standing Cobb angle. If so, the bending Cobb angle and the lateral angle are sequentially subjected to a first classification condition and a second classification condition. If both the first and second classification conditions are met, the curvature type is determined to be Type 4. If the first classification condition is met but the second classification condition is not met, the curvature type is determined to be Type 2. If the first classification condition is not met but the second classification condition is met, the curvature type is determined to be Type 3. If neither the first classification condition nor the second classification condition is met, the curvature type is determined to be Type 1. If not, the bending Cobb angle and the lateral angle are subjected to a third classification condition. If the third classification condition is met, the curvature type is determined to be Type 6. If the third classification condition is not met, the curvature type is determined to be Type 5.

[0072] Among them, the first classification condition is: the Cobb angle of the upper thoracic curve in the bending position is not less than 25° or the first lateral angle is not less than 20°; the second classification condition is: the Cobb angle of the lumbar curve in the bending position is not less than 25° or the second lateral angle is not less than 20°; the third classification condition is: the Cobb angle of the main thoracic curve in the bending position is not less than 25° or the second lateral angle is not less than 20°.

[0073] For example, the classification flow chart of the lumbar revision type is as follows: Figure 8 As shown, Figure 8This is a flow chart for lumbar vertebrae classification according to an embodiment of the present invention. In this embodiment, the second classification unit is used to determine the central sacral vertical line based on a standing spinal X-ray film, and to determine the two pedicles of the apex lumbar curve based on the coordinates of the key points of each vertebra in the preprocessed image corresponding to the standing spinal X-ray film. If the central sacral vertical line is between the two pedicles, the lumbar vertebrae classification is type A; if the central sacral vertical line is on either pedicle, the lumbar vertebrae classification is type B; if the central sacral vertical line is outside the two pedicles, the lumbar vertebrae classification is type C.

[0074] Specifically, the Central Sacral Vertical Line (CSVL) is a line extending vertically upward from the center of the sacrum and is commonly used to assess the overall balance of the spine. When performing a revised lumbar classification, the two vertebral angles at the upper edge of the first sacrum are first located based on a standing spinal X-ray. The average of the horizontal coordinates of these two vertebral angles is then used as S1, the midpoint of the upper edge of the first sacrum. A line extending vertically upward through this point, parallel to the edge of the X-ray film, is the central sacral perpendicular line. This perpendicular line is not perpendicular to the upper edge of S1, but to the lower edge of the X-ray film, i.e., the line with X = the horizontal coordinate of the midpoint of S1. The coordinates of the left and right midpoints of 17 vertebrae are then obtained based on the coordinates of the key points of each vertebra in the preprocessed image corresponding to the standing spinal X-ray film. A curve of the coordinates of the 17 vertebral midpoints is fitted using a fifth-order function. The solution of [13-17] where the first-order derivative of this curve is zero is the apex vertebra of the lumbar curvature. After the apical vertebra is identified, an image of the cone can be captured based on the coordinates of the vertebra's four key points and fed into a U-Net image segmentation network for pedicle detection. U-Net employs an encoder-decoder architecture that downsamples the input image, extracts deep features through convolution and pooling, and gradually upsamples the image to restore its spatial resolution. Using skip connections, the encoder's feature map information is passed to the decoder, facilitating more precise boundary localization. The final output is a prediction mask the same size as the input image, where the value of each pixel indicates whether it belongs to a pedicle. Finally, the positional relationship between the central sacral vertical line and the pair of pedicles detected at the apical vertebra of the lumbar curve is used to determine the correct lumbar spine type.

[0075] See Figure 9 , Figure 9 This is an example diagram of the relationship between CSVL and pedicle provided by an embodiment of the present invention. Figure 9As shown in the figure, assume that a pair of pedicles A and B of the detected apical vertebra of lumbar curve are provided, where the range of pedicle A on the X-axis is (X1, X2), the range of pedicle B on the X-axis is (X3, X4), and CSVL is represented as X0 on the X-axis. Then the relationship between CSVL and the pedicles includes: CSVL is on any one of the pedicles, that is, X1 ≤ X0 ≤ X2 or X3 ≤ X0 ≤ X4; CSVL is between the two pedicles, that is, X2 < X0 < X3. CSVL is outside the pedicles, that is, X0 < X1 or X0 > X4. According to the relationship between CSVL and the pedicles, the corresponding classification of the lumbar modification type can be determined as Figure 10 as shown in the figure Figure 10 is a schematic diagram of the classification of the lumbar modification type provided by an embodiment of the present invention

[0076] Exemplarily, the classification flow chart of thoracic sagittal kyphosis is as Figure 11 shown in the figure Figure 11 is a classification flow chart of thoracic sagittal kyphosis provided by an embodiment of the present invention. In this embodiment, the third classification unit is used to determine whether the third lateral angle is less than 10° according to the lateral angle. If so, the thoracic sagittal kyphosis is classified as negative type. If not, it is determined whether the third lateral angle is less than 40°. If so, the thoracic sagittal kyphosis is classified as normal type. If not, the thoracic sagittal kyphosis is classified as positive type

[0077] The adolescent idiopathic scoliosis auxiliary diagnosis and evaluation system based on X-ray films in the embodiments of the present invention realizes the key point detection of standing position, Bending position and lateral spine X-ray films, automatic Cobb angle measurement and Lenke classification, including bending type, lumbar modification type and thoracic sagittal kyphosis classification, significantly improves the accuracy, efficiency and reliability of auxiliary diagnosis, and provides comprehensive support for the clinical diagnosis and treatment decision-making of scoliosis. Through deep learning and key point detection technology, it can accurately identify the vertebral key points in various positions including standing position, left and right bending positions and lateral position, so as to calculate a more accurate Cobb angle, and significantly improve the accuracy of the Cobb angle measurement result

[0078] Furthermore, please refer to Figure 12 , Figure 12 which is a structural block diagram of another adolescent idiopathic scoliosis auxiliary diagnosis and evaluation system based on X-ray films provided by an embodiment of the present invention, as Figure 12 shown in the figure, the adolescent idiopathic scoliosis auxiliary diagnosis and evaluation system based on X-ray films in this embodiment further includes a visualization module, and the visualization module is used to display the standing Cobb angle, Bending Cobb angle, lateral angle and Lenke classification result

[0079] In this embodiment, visualization software can be used to visualize the standing Cobb angle, bending Cobb angle, lateral angle, and Lenke classification results. Figure 13-15 , Figure 13 This is a diagram of a Cobb angle measurement interface for a standing position provided by an embodiment of the present invention; Figure 14 This is a diagram of a measurement interface for a bending position Cobb angle provided by an embodiment of the present invention; Figure 15 This is an interface diagram of lateral angles and Lenke classification provided by an embodiment of the present invention. In the visualization processing, different colors can be used to distinguish the Cobb angles according to their size and the vertebra where they are located. This allows the doctor to see at a glance which two vertebrae correspond to the angle between the Cobb angles. At the same time, the calculated Cobb angles of the upper thoracic curve, main thoracic curve and thoracolumbar curve / lumbar curve are also output in the software interface so that the doctor can clearly understand the values ​​of these angles. In addition, in order to better guide clinical judgment, a grade classification of the Cobb angle is also given to help doctors assess the severity of scoliosis and thus select the appropriate treatment plan. Through the above visualization processing, the angle of scoliosis and the Lenke classification can be intuitively understood, further providing support and reference for the doctor's diagnosis and treatment.

[0080] See Figure 16 , Figure 16 This is an interactive flow chart of a visualization software provided by an embodiment of the present invention, such as Figure 16 As shown, in the Standing Cobb Angle Measurement interface, enter the standing spinal X-ray to be tested. Click Detect to view the Cobb angle in the results box. To further measure the Cobb angle or lateral angle of the bending spinal X-ray, simply click the Bending Position Detect or Lateral Position Detect button. In the pop-up measurement interface, enter the left and right bending spinal X-rays or lateral spinal X-rays, respectively. Click Detect to obtain the left and right bending Cobb angles or the three lateral angles. Finally, click Lenke Classification, which automatically combines the standing and bending Cobb angles and the three lateral angles to determine the classification result. This software allows doctors to more conveniently measure scoliosis angles, simplifying workflows and improving efficiency and accuracy. It also assists doctors in quickly performing Lenke classification. The software can also be expanded and optimized based on actual needs to meet the diverse needs of doctors.

[0081] It should be noted that, in this document, relational terms such as first and second are used solely to distinguish one entity or operation from another, and do not necessarily require or imply any actual relationship or order between these entities or operations. Furthermore, the terms "comprise," "include," or any other variations thereof are intended to encompass non-exclusive inclusion, such that an article or device comprising a list of elements includes not only those elements but also other elements not explicitly listed. Without further limitation, an element defined by the phrase "comprising a..." does not preclude the presence of additional identical elements in the article or device comprising the element. Terms such as "connected" or "connected" are not limited to physical or mechanical connections but may include electrical connections, whether direct or indirect. References to orientations or positional relationships, such as "upper," "lower," "left," and "right," are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely to facilitate description and simplify the description of the present invention. They do not indicate or imply that the device or element referred to must have, be constructed, or operate in a specific orientation, and are therefore not to be construed as limiting the present invention.

[0082] The above is a further detailed description of the present invention in conjunction with specific preferred embodiments, and the specific implementation of the present invention should not be considered to be limited to these descriptions. For those skilled in the art to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, and all of these should be considered to fall within the scope of protection of the present invention.

Claims

1. An X-ray-based auxiliary diagnosis and evaluation system for adolescent idiopathic scoliosis, characterized in that: include: An image preprocessing module is used to perform ROI area detection on the spinal X-ray film to be detected, obtain a cut-off image containing the spine, and perform size adjustment and normalization processing on the cut-off image to obtain a preprocessed image; A key point detection module, used to detect the pre-processed image using a preset trained key point detection model to obtain the coordinates of the key points of each vertebra in the pre-processed image; A Cobb angle measurement module, used for calculating the standing Cobb angle, the bending Cobb angle and the lateral angle according to the coordinates of the key points of each vertebra in the preprocessed image; The Lenke classification module is used to perform Lenke classification according to the spinal X-ray film to be detected, the standing Cobb angle, the bending Cobb angle and the lateral angle to obtain a Lenke classification result.

2. The X-ray-based adolescent idiopathic scoliosis auxiliary diagnosis and evaluation system according to claim 1 is characterized in that: The spinal X-ray films to be tested include: spinal X-ray films in standing position, spinal X-ray films in left bending position, spinal X-ray films in right bending position and spinal X-ray films in lateral position.

3. The X-ray-based adolescent idiopathic scoliosis auxiliary diagnosis and evaluation system according to claim 1, characterized in that: The key point detection model is a key point detection model based on heat map regression that introduces a channel attention mechanism.

4. The X-ray-based adolescent idiopathic scoliosis auxiliary diagnosis and evaluation system according to claim 1, characterized in that: The key points include the upper left point, the upper right point, the lower left point and the lower right point of the vertebra.

5. The X-ray-based adolescent idiopathic scoliosis auxiliary diagnosis and evaluation system according to claim 2, characterized in that: The Cobb angle measurement module comprises: A standing position Cobb angle measurement unit, used to determine a first tilt vector of each vertebra according to the coordinates of a key point of each vertebra in a preprocessed image corresponding to the standing position spinal X-ray film, calculate the angle between two vertebrae with the largest angle among all vertebrae according to the first tilt vector as the standing position Cobb1 angle, and calculate the standing position Cobb2 angle and the standing position Cobb3 angle according to the type of scoliosis in the spinal X-ray film to be detected and the position of the vertebra corresponding to the standing position Cobb1 angle; A bending position Cobb angle measurement unit, used for calculating the bending position Cobb1 angle, the bending position Cobb2 angle and the bending position Cobb3 angle according to the coordinates of the key points of each vertebra in the preprocessed image corresponding to the spinal X-ray film of the left bending position, the coordinates of the key points of each vertebra in the preprocessed image corresponding to the spinal X-ray film of the right bending position, and the vertebral positions corresponding to the standing position Cobb1 angle, the standing position Cobb2 angle and the standing position Cobb3 angle; The lateral angle measurement unit is used to determine the second inclination vector of each vertebra according to the coordinates of the key points of each vertebra in the preprocessed image corresponding to the lateral spinal X-ray film, and calculate the angle between vertebra T2 and vertebra T5 as the first lateral angle, the angle between vertebra T10 and vertebra L2 as the second lateral angle, and the angle between vertebra T5 and vertebra T12 as the third lateral angle according to the second inclination vector.

6. The X-ray-based adolescent idiopathic scoliosis auxiliary diagnosis and evaluation system according to claim 5, characterized in that: According to the scoliosis type of the spinal X-ray film to be detected and the vertebral position corresponding to the standing Cobb1 angle, the standing Cobb2 angle and the standing Cobb3 angle are calculated, including: Determine the type of scoliosis of the spinal X-ray film to be detected according to the coordinates of the key points of each vertebra in the preprocessed image corresponding to the spinal X-ray film in the standing position; If the scoliosis type is type C, then according to the first tilt vector, the angle between the upper vertebra and vertebra T1 of the standing Cobb1 angle is calculated as the standing Cobb2 angle, and the angle between the lower vertebra and vertebra L5 of the standing Cobb1 angle is calculated as the standing Cobb3 angle; If the scoliosis type is S-type, determining whether the standing Cobb1 angle is an upper thoracic curve; If the standing position Cobb1 angle is an upper thoracic curve, then according to the first inclination vector, the maximum angle between the lower vertebra of the standing position Cobb1 angle and all the vertebrae below the lower vertebra of the standing position Cobb1 angle is calculated as the standing position Cobb2 angle, and the maximum angle between the lower vertebra of the standing position Cobb2 angle and all the vertebrae below the lower vertebra of the standing position Cobb2 angle is calculated as the standing position Cobb3 angle; If the standing Cobb1 angle is not the upper thoracic curve, then according to the first inclination vector, the maximum angle between the upper vertebra of the standing Cobb1 angle and all the vertebrae above the upper vertebra of the standing Cobb1 angle is calculated as the standing Cobb2 angle, and the maximum angle between the lower vertebra of the standing Cobb1 angle and all the vertebrae below the lower vertebra of the standing Cobb1 angle is calculated as the standing Cobb3 angle.

7. The X-ray-based adolescent idiopathic scoliosis auxiliary diagnosis and evaluation system according to claim 5, characterized in that: According to the coordinates of the key points of each vertebra in the preprocessed image corresponding to the spinal X-ray film of the left bending position, the coordinates of the key points of each vertebra in the preprocessed image corresponding to the spinal X-ray film of the right bending position, and the vertebral positions corresponding to the standing position Cobb1 angle, the standing position Cobb2 angle, and the standing position Cobb3 angle, the bending position Cobb1 angle, the bending position Cobb2 angle, and the bending position Cobb3 angle are calculated, including: Determine the third tilt vector of each vertebra according to the coordinates of the key points of each vertebra in the preprocessed image corresponding to the spinal X-ray film of the left bending position, and calculate the left bending position Cobb1 angle, the left bending position Cobb2 angle and the left bending position Cobb3 angle according to the vertebral positions corresponding to the standing position Cobb1 angle, the standing position Cobb2 angle and the standing position Cobb3 angle using the third tilt vector; Determine the fourth inclination vector of each vertebra according to the coordinates of the key points of each vertebra in the preprocessed image corresponding to the spinal X-ray film in the right bending position, and calculate the right bending position Cobb1 angle, the right bending position Cobb2 angle and the right bending position Cobb3 angle according to the vertebral positions corresponding to the standing position Cobb1 angle, the standing position Cobb2 angle and the standing position Cobb3 angle using the fourth inclination vector; The smaller value between the left bending position Cobb1 angle and the right bending position Cobb1 angle is selected as the Bending position Cobb1 angle, the smaller value between the left bending position Cobb2 angle and the right bending position Cobb2 angle is selected as the Bending position Cobb2 angle, and the smaller value between the left bending position Cobb3 angle and the right bending position Cobb3 angle is selected as the Bending position Cobb3 angle.

8. The X-ray-based adolescent idiopathic scoliosis auxiliary diagnosis and evaluation system according to claim 5, characterized in that: The Lenke classification module includes: A first classification unit is used to determine whether the Cobb angle of the main thoracic curve in the standing position is the largest according to the Cobb angle in the standing position. If so, the first classification condition and the second classification condition are judged on the Bending Cobb angle and the lateral angle in sequence. If both the first classification condition and the second classification condition are satisfied, the bending type is determined to be Type 4. If the first classification condition is satisfied and the second classification condition is not satisfied, the bending type is determined to be Type 2. If the first classification condition is not satisfied and the second classification condition is satisfied, the bending type is determined to be Type 3. If both the first classification condition and the second classification condition are not satisfied, the bending type is determined to be Type 1. If not, the third classification condition is judged on the Bending Cobb angle and the lateral angle. If the third classification condition is satisfied, the bending type is determined to be Type 6. If the third classification condition is not satisfied, the bending type is determined to be Type 5. The second classification unit is used to determine the central sacral vertical line according to the standing spinal X-ray film, and determine the two pedicles of the apical vertebra of the lumbar curve according to the coordinates of the key points of each vertebra in the preprocessed image corresponding to the standing spinal X-ray film, if the central sacral vertical line is between the two pedicles, the lumbar vertebrae are modified type A, if the central sacral vertical line is on any pedicle, the lumbar vertebrae are modified type B, if the central sacral vertical line is outside the two pedicles, the lumbar vertebrae are modified type C; The third classification unit is used to determine whether the third lateral angle is less than 10° based on the lateral angle. If so, the thoracic sagittal kyphosis classification is negative. If not, it is determined whether the third lateral angle is less than 40°. If so, the thoracic sagittal kyphosis classification is normal. If not, the thoracic sagittal kyphosis classification is positive.

9. The X-ray-based adolescent idiopathic scoliosis auxiliary diagnosis and evaluation system according to claim 8, characterized in that: The first classification condition is: the Cobb angle of the thoracic curve in the bending position is not less than 25° or the first lateral angle is not less than 20°; The second classification condition is: the Cobb angle of the bending lumbar curve is not less than 25° or the second lateral angle is not less than 20°; The third classification condition is that the Cobb angle of the main thoracic curve in the bending position is not less than 25° or the second lateral angle is not less than 20°.

10. The X-ray-based adolescent idiopathic scoliosis auxiliary diagnosis and evaluation system according to claim 1, characterized in that: The X-ray-based adolescent idiopathic scoliosis auxiliary diagnosis and evaluation system also includes a visualization module for displaying the standing Cobb angle, the bending Cobb angle, the lateral angle and the Lenke classification result.

Images