Intervertebral disc positioning method and device, electronic device and storage medium

Abstract

Embodiments of the present specification provide an intervertebral disc positioning method and device, electronic equipment and storage medium, the method comprising: obtaining a medical image of a target site, preprocessing the medical image to obtain a target image, inputting the target image into a trained target detection model to obtain a cone region in the target image, confirming a target vertebral region and a neighboring vertebral apex of a neighboring vertebra in the target image based on the vertebral region, confirming an adjacent edge edge pixel of the neighboring vertebra according to the target vertebral region and the neighboring cone apex, and confirming an intervertebral disc positioning result of the neighboring vertebra based on the adjacent edge edge pixel. The intervertebral disc positioning method improves the efficiency of target detection by preprocessing the medical image, i.e. improves the efficiency of the confirmed vertebral region, and then improves the accuracy and robustness of intervertebral disc positioning by confirming the adjacent edge edge pixel of the neighboring vertebra in the vertebral region.

Description

Technical Field

[0001] This application relates to the field of medical image processing technology, and in particular to intervertebral disc localization methods, devices, electronic devices and storage media. Background Technology

[0002] Intervertebral disc diseases are very common clinical conditions, including disc herniation and disc degeneration. These conditions usually require imaging examinations for diagnosis. The procedure for intervertebral disc scanning typically includes a first sagittal plane scan, spinal confirmation, disc localization, and transverse scanning. Based on the obtained sagittal plane images, the technician will mark the disc location and establish positioning lines to help determine the disc's precise location and orientation. Then, guided by these positioning lines, a transverse scan is performed to obtain more detailed and accurate images of the disc structure.

[0003] However, this method requires experienced technicians to manually label and analyze the data, which is not only time-consuming and labor-intensive, but also subject to subjectivity and error. Summary of the Invention

[0004] The main objective of this application is to provide a method, device, electronic device, and storage medium for intervertebral disc localization, aiming to achieve intelligent intervertebral disc localization and improve its accuracy and robustness. The technical solution is as follows:

[0005] Firstly, embodiments of this specification provide a method for locating an intervertebral disc, including:

[0006] Acquire medical images of the target area; the target area is any part of the spine;

[0007] The medical images are preprocessed to obtain the target image;

[0008] The target image is input into a trained target detection model to obtain the cone region in the target image;

[0009] Based on the vertebral region, the target vertebral region and the adjacent vertebral apex in the target image are identified;

[0010] Based on the target vertebral region and the adjacent vertebral vertices, the adjacent edge pixels of the adjacent vertebral bodies are confirmed;

[0011] The intervertebral disc localization result of the adjacent vertebrae is confirmed based on the adjacent edge pixels.

[0012] Secondly, embodiments of this specification provide an intervertebral disc positioning device, comprising:

[0013] The acquisition module is used to acquire medical images of the target area; the target area is any part of the spine.

[0014] A preprocessing module is used to preprocess the medical image to obtain the target image;

[0015] The model detection module is used to input the target image into a trained target detection model to obtain the cone region in the target image;

[0016] The adjacent vertebral body localization module is used to identify the target vertebral body region and the adjacent vertebral body apex in the target image based on the vertebral body region.

[0017] The vertebral body edge confirmation module is used to confirm the adjacent edge pixels of the adjacent vertebrae based on the target vertebral body region and the adjacent vertebral vertebrae vertebrae vertebrae.

[0018] The positioning result confirmation module is used to confirm the intervertebral disc positioning result of the adjacent vertebrae based on the adjacent edge pixels.

[0019] Thirdly, embodiments of this specification provide an electronic device, the electronic device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the method described above.

[0020] Fourthly, embodiments of this specification provide a storage medium storing a computer program, which, when executed by a processor, implements the steps of the method described above.

[0021] In the embodiments of this specification, a medical image of any target location in the spine is acquired, preprocessed to obtain a target image, and input into a trained target detection model to obtain the vertebral region within the target image. Based on the vertebral region, the target vertebral region and adjacent vertebral vertebrae in the target image are identified. Based on the target vertebral region and adjacent vertebral vertebrae vertebrae, the adjacency edge pixels of adjacent vertebrae are identified, and the intervertebral disc localization result of adjacent vertebrae is confirmed based on the adjacency edge pixels. This intervertebral disc localization method combines the target detection model to locate the vertebral region, and then performs intervertebral disc localization based on the adjacency edges of adjacent vertebrae identified within the vertebral region. Using adjacency edges to characterize the vertebral orientation provides high robustness, thereby improving the robustness of intervertebral disc localization. Simultaneously, this method also improves the accuracy and precision of intervertebral disc detection. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments or prior art of this specification, the drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a schematic diagram illustrating the application of an intervertebral disc localization method provided in the embodiments of this specification;

[0024] Figure 2 This is a flowchart illustrating a method for locating an intervertebral disc provided in an embodiment of this specification;

[0025] Figure 3 This is a schematic diagram illustrating an example of an intervertebral disc localization method provided in the embodiments of this specification;

[0026] Figure 4 This is a schematic diagram illustrating an example of an intervertebral disc localization method provided in the embodiments of this specification;

[0027] Figure 5 This is a flowchart illustrating a method for locating an intervertebral disc provided in an embodiment of this specification;

[0028] Figure 6 This is a schematic diagram illustrating an example of an intervertebral disc localization method provided in the embodiments of this specification;

[0029] Figure 7 This is a schematic diagram illustrating an example of an intervertebral disc localization method provided in the embodiments of this specification;

[0030] Figure 8 This is a flowchart illustrating a method for locating an intervertebral disc provided in an embodiment of this specification;

[0031] Figure 9 This is a schematic diagram illustrating an example of an intervertebral disc localization method provided in the embodiments of this specification;

[0032] Figure 10 This is a schematic diagram illustrating an example of an intervertebral disc localization method provided in the embodiments of this specification;

[0033] Figure 11 This is a schematic diagram of the structure of an intervertebral disc positioning device provided in the embodiments of this specification;

[0034] Figure 12 This is a schematic diagram of the structure of an electronic device provided in the embodiments of this specification. Detailed Implementation

[0035] The technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this specification.

[0036] In the description of this specification, it should be understood that the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. In the description of this specification, it should be noted that, unless otherwise expressly specified and limited, "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices. Those skilled in the art can understand the specific meaning of the above terms in this specification based on the specific circumstances. Furthermore, in the description of this specification, unless otherwise stated, "multiple" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship.

[0037] The intervertebral disc localization device provided in the embodiments of this specification can be a terminal device such as a mobile phone, computer, or tablet computer, or it can be a module in a terminal device used to implement the intervertebral disc localization method. The intervertebral disc localization device can acquire medical images of the target area, preprocess the medical images to obtain target images, input the target images into a trained target detection model to obtain the vertebral region in the target image, identify the target vertebral region and adjacent vertebral vertebrae in the target image based on the vertebral region, identify the adjacent vertebral edge pixels of adjacent vertebrae based on the target vertebral region and adjacent vertebral vertebrae, and confirm the intervertebral disc localization result of adjacent vertebrae based on the adjacent edge pixels.

[0038] Please see also Figure 1 This diagram illustrates the application of an intervertebral disc localization method as provided in the embodiments of this specification. Please refer to... Figure 1In the embodiments of this specification, the intervertebral disc positioning device can be connected to medical imaging equipment via wired or wireless means. The medical imaging equipment acquires medical images of the target area, and the intervertebral disc positioning device preprocesses the medical images to obtain the target image. The target image is then input into a trained target detection model to obtain the vertebral region in the target image. Based on the vertebral region, the target vertebral region and adjacent vertebral vertebrae in the target image are identified. Based on the target vertebral region and adjacent vertebral vertebrae vertebrae vertebrae vertebrae vertebrae, the adjacent edge pixels of adjacent vertebrae are identified. Based on the adjacent edge pixels, the intervertebral disc positioning result of adjacent vertebrae is confirmed.

[0039] The intervertebral disc localization method provided in this specification will be described in detail below with reference to specific embodiments.

[0040] Please see Figure 2 This is a flowchart illustrating a method for locating an intervertebral disc, as provided in the embodiments of this specification. Figure 2 As shown, the method described in the embodiments of this specification may include the following steps S101-S106.

[0041] S101, Acquire medical images of the target area;

[0042] In one embodiment, when intervertebral disc localization is required at a target location, medical images of the target location need to be acquired. Specifically, medical imaging equipment can be used to acquire these images. This equipment can be an MR (Magnetic Resonance) scanner or a CT (Computed Tomography) scanner, etc. The target location is any part of the spine. The spine can be divided into five parts from top to bottom: cervical, thoracic, lumbar, sacral, and coccygeal vertebrae. After determining the target location to be examined, a scan is performed using the medical imaging equipment, following preset scanning parameters, to acquire medical images of the area to be examined. Depending on the medical imaging equipment, the acquired medical images can be CT images, MRI images, etc. The medical images can be sagittal images including the target location or coronal images including the target location. The sagittal plane refers to the plane that divides the human body into two symmetrical parts, left and right; a sagittal image refers to an image scanned parallel to the sagittal plane. The coronal plane refers to the plane that divides the human body into two symmetrical parts, anterior and posterior; a coronal image refers to an image scanned parallel to the coronal plane.

[0043] S102, The medical image is preprocessed to obtain the target image;

[0044] In one embodiment, after acquiring a medical image, it can be preprocessed to improve image quality. The preprocessed image yields the target image, facilitating subsequent analysis and localization. It is understood that medical images are typically high dynamic range (DMU) DICOM images. Since the main observation area of ​​medical images containing cones is distributed in the low to medium grayscale region, in a feasible implementation, dynamic range compression can be used as a preprocessing method to improve image contrast and cone detection efficiency. Dynamic range compression methods can include histogram statistics, brightness range search, global mapping, deep learning methods, etc.

[0045] Optionally, since the target detection model has size requirements for the input image, the medical image can be processed to a preset size during preprocessing to obtain the target image.

[0046] S103, input the target image into the trained target detection model to obtain the cone region in the target image;

[0047] Understandably, target images often include other human tissues, but these are not the areas of interest. Therefore, a pre-trained object detection model can be used to identify vertebrae in the target image to confirm the location of the vertebrae. The vertebral region includes the areas containing the cervical, thoracic, lumbar, sacral, and coccygeal vertebrae.

[0048] The object detection model can be an R-CNN series model, such as the R-CNN model or the Faster R-CNN model, or a YOLO model, such as the YOLOv5 model or the YOLOv8 model. Taking the YOLOv8 model as an example, multiple spine images for model training can be pre-acquired; the spine images are manually labeled to obtain the vertebral regions and their corresponding area ranges; and the YOLOv8 model to be trained is then trained based on the spine images and the corresponding labeling results to obtain a trained YOLOv8 model. Optionally, to ensure the accuracy of the vertebral detection results, an upper limit on the number of pixels for a single vertebra is set based on experience. In the detection results (vertebral regions) output by the object detection model, a region growing algorithm is used to expand the vertebral regions to obtain the final vertebral regions.

[0049] S104, Based on the vertebral region, confirm the target vertebral region and the adjacent vertebral apex in the target image;

[0050] In one embodiment, the vertebral body region may include multiple vertebrae. After identifying the vertebral body region, any adjacent vertebrae can be identified as adjacent vertebrae, or specific adjacent vertebrae to be observed can be specified, and their corresponding target vertebral body region can be obtained. Adjacent vertebrae in the spine refer to two adjacent vertebrae of the vertebral column, connected by intervertebral discs and joints. The target vertebral body region is the vertebral body region containing the adjacent vertebrae, and the vertices of the adjacent vertebrae are the vertices adjacent to the intervertebral disc portion between each vertebra and its neighboring vertebrae. Optionally, the output vertebral body region may already have the vertex detection results of each vertebrae marked. If the output vertebral body region does not have the vertex detection results marked, vertex identification can be performed on the vertebrae in the target vertebral body region after identifying the target vertebral body region. For example, a corner detection algorithm can be used to obtain the vertices of adjacent vertebrae (all vertices of adjacent vertebrae), and the adjacent vertices can be identified from all the vertices of adjacent vertebrae.

[0051] S105, Based on the target vertebral region and the adjacent vertebral vertices, confirm the adjacent edge pixels of the adjacent vertebral bodies;

[0052] In one embodiment, the adjacent edge pixels of the adjacent vertebrae are determined based on the identified target vertebral region and the vertices of the adjacent vertebrae. It is understood that the adjacent edge of an adjacent vertebra is the two adjacent edges between two adjacent vertebrae in the target vertebral region. In a feasible implementation, the target vertebral region may include the area where the vertebra is located and the edge contour of that area, so the adjacent edge pixels of the adjacent vertebrae can be determined from the edge contour.

[0053] Please see Figure 3 , Figure 3 This specification provides an example of an intervertebral disc localization method. Optionally, after identifying the vertebral body region from the target image, polygon fitting is performed on each segmented vertebral body to obtain a polygonal vertebral body. The vertex positions of the polygonal vertebral bodies are then identified, resulting in the vertebral body region and the vertebral vertebrae. By selecting adjacent vertebrae, adjacent vertebral body regions within the vertebral body region are further identified, i.e. Figure 3 Vertices 1-4 in the diagram are adjacent vertices of the vertices. The adjacent edge pixels of the adjacent vertices are identified based on the adjacent vertices and the vertices region.

[0054] S106, Based on the adjacent edge pixels, confirm the intervertebral disc localization result of the adjacent vertebrae.

[0055] In one embodiment, the intervertebral disc location of adjacent vertebrae is determined based on the identified edge pixels of the adjacent edges. It is understood that the intervertebral disc is the middle portion of two adjacent vertebrae; the orientation and position of the two vertebrae can be determined based on their adjacent edges, and the location of the intervertebral disc can be determined based on the two adjacent edges.

[0056] In one feasible implementation, the centroid and principal vector direction of the adjacent edge pixels can be identified, and the intervertebral disc positioning line of the adjacent vertebral body can be identified based on the centroid and principal vector direction, and the intervertebral disc positioning line can be identified as the intervertebral disc positioning result.

[0057] For example, in the spine, vertebral bodies L5 and S1 are at the distal end. In medical images, these two vertebrae may be incompletely captured or compressed, which may lead to large errors in vertex extraction. If the intervertebral disc localization auxiliary line is located based on the vertebral body vertex, the error will also be large, resulting in inaccurate intervertebral disc localization. However, using the edge pixel distribution of the adjacent edges of adjacent vertebrae can basically depict the approximate direction of the vertebrae, and its direction depiction accuracy is higher. Using the edge pixels of the adjacent edges of two adjacent vertebrae to extract the principal vector can improve the robustness of intervertebral disc localization.

[0058] Please see Figure 4 , Figure 4 This specification provides an example schematic diagram of an intervertebral disc localization method, wherein the solid line L1 represents the intervertebral disc localization line identified by the vertebral apex of the adjacent vertebral body, and the dashed line L2 represents the intervertebral disc localization line identified by the edge pixels of the adjacent vertebral bodies.

[0059] Optionally, after obtaining the intervertebral disc localization results, a cross-sectional scan of the intervertebral disc can be performed based on the selected intervertebral disc and the corresponding intervertebral disc localization results.

[0060] In the embodiments of this specification, medical images of the target area are acquired, preprocessed to obtain a target image, and input into a trained target detection model to obtain the vertebral region in the target image. Based on the vertebral region, the target vertebral region and adjacent vertebral vertebrae in the target image are identified. Based on the target vertebral region and adjacent vertebral vertebrae vertebrae vertebrae vertebrae, the adjacency edge pixels of adjacent vertebrae are identified. Based on the adjacency edge pixels, the intervertebral disc localization result of adjacent vertebrae is confirmed. The target detection model efficiently locates the vertebral region, and the robustness of the vertebral direction is characterized by the adjacency edges of adjacent vertebrae, thereby improving the robustness and accuracy of intervertebral disc localization.

[0061] Please see Figure 5 This is a flowchart illustrating a method for locating an intervertebral disc provided in an embodiment of this specification. Figure 5 As shown, the method described in the embodiments of this specification may include the following steps S201-S202.

[0062] S201, Obtain the dynamic range compression parameters corresponding to the target part;

[0063] In one embodiment, dynamic range compression can be used to enhance image contrast and improve cone detection efficiency during medical image preprocessing. Specifically, dynamic range compression parameters corresponding to the target location can be obtained, including an upper and lower dynamic range limit. Since different parts of the spine, such as the cervical, thoracic, and lumbar vertebrae, contain different tissues in their image fields of view, using dynamic range compression parameters corresponding to the target location can improve image processing performance. In a feasible implementation, a set of dynamic range compression parameters corresponding to various parts of the spine can be prepared in advance. Once the target location of the current medical image is identified, the dynamic range compression parameters corresponding to the target location can be determined from the set of dynamic range compression parameters.

[0064] Furthermore, in one embodiment, before obtaining the dynamic range compression parameters corresponding to the target region, the method further includes:

[0065] S2011, Obtain a set of historical images of the target area, and adjust the dynamic range of each historical image in the set of historical images to the optimal observation range;

[0066] Specifically, a historical image set for the target area can be pre-acquired. This historical image set includes medical images acquired for the target area, and different observation angles can be selected to obtain a richer collection of historical images. After acquiring the historical image set, each historical image can be adjusted. For example, histogram adjustments, exposure adjustments, and contrast adjustments can be used to optimize the historical images for observation. The dynamic range of the adjusted historical images is the optimal observation range. It's understandable that medical images are typically MRI images, so the background is nearly black. The lower limit is mainly used to filter background pixels and reduce interference.

[0067] S2012, confirm the mean of the optimal observation range, confirm the upper limit of the mean as the upper limit of the dynamic range of the target part, and confirm the lower limit of the mean as the lower limit of the dynamic range of the target part.

[0068] Specifically, the mean of the optimal observation range of each historical image is calculated to obtain a mean optimal observation range. The upper limit of the mean is determined as the upper limit of the dynamic range of the target part, and the lower limit of the mean is determined as the lower limit of the dynamic range of the target part. Optionally, the mean of the upper limits of the dynamic range of the historical images included in the historical image set can be directly determined as the lower limit of the dynamic range of the target part, and the mean of the lower limits of the dynamic range of the included historical images can be determined as the upper limit of the dynamic range of the target part.

[0069] S2013, The upper limit of the dynamic range and the lower limit of the dynamic range are confirmed as the dynamic range compression parameters of the target part.

[0070] Specifically, after obtaining the upper limit and lower limit of the dynamic range, the dynamic range compression parameters of the target part are determined.

[0071] S202, adjust the pixels in the medical image that are above the upper limit of the dynamic range to the upper limit of the dynamic range, and adjust the pixels in the medical image that are below the lower limit of the dynamic range to the lower limit of the dynamic range, so as to obtain the target image.

[0072] In one embodiment, processing a medical image using a dynamic range parameter may include adjusting all pixels in the original medical image whose pixel values ​​are higher than the upper limit of the dynamic range to the upper limit of the dynamic range, and adjusting all pixels in the medical image whose pixel values ​​are lower than the lower limit of the dynamic range to the lower limit of the dynamic range.

[0073] Please see Figure 6 , Figure 6 This is a schematic diagram illustrating an example of an intervertebral disc localization method provided in the embodiments of this specification. It shows spinal images before and after dynamic range compression processing. The image before processing is a medical image, and the image after processing is the target image.

[0074] Further, in one embodiment, adjusting pixels in the medical image that are above the upper limit of the dynamic range to the upper limit of the dynamic range, and adjusting pixels in the medical image that are below the lower limit of the dynamic range to the lower limit of the dynamic range, to obtain the target image, may include the following steps S2021-S2022:

[0075] S2021, adjust the pixels in the medical image that are above the upper limit of the dynamic range to the upper limit of the dynamic range, and adjust the pixels in the medical image that are below the lower limit of the dynamic range to the lower limit of the dynamic range, so as to obtain a medical image after dynamic range compression processing;

[0076] Specifically, dynamic range compressed medical images can be obtained first through dynamic range compression.

[0077] S2022, The medical image after dynamic range compression is subjected to nonlinear pixel value conversion to obtain the target image.

[0078] Specifically, after dynamic range compression, the pixel values ​​of the image can be mapped to the range of 0-255 to obtain the target image. During the mapping process, gamma transformation is used to improve image contrast and increase cone detection efficiency. Gamma transformation is a non-linear operation, and its main purpose is to enhance the data distribution of the target domain through grayscale value mapping, thereby achieving contrast enhancement. Gamma transformation can achieve more flexible transformations depending on the parameters, and its grayscale transformation function is defined as follows:

[0079] s = cr γ

[0080] Applying gamma transform to dynamically range compressed spinal images can improve the contrast between the vertebrae and intervertebral discs, thereby increasing the efficiency of limb detection. The effect is as follows: Figure 7 , Figure 7 This is a schematic diagram illustrating an example of an intervertebral disc localization method provided in the embodiments of this specification, showing a comparison before and after nonlinear pixel value conversion. The grayscale image is also the medical image after dynamic range compression, and γ = 0.7 corresponds to the target image after gamma transformation.

[0081] In the embodiments of this specification, by obtaining the dynamic range compression parameters corresponding to the target area, and based on the upper and lower limits of the dynamic range parameters, pixels in the medical image above the upper limit are adjusted to the upper limit, and pixels below the lower limit are adjusted to the lower limit, thus obtaining the target image. Dynamic range compression improves image contrast and cone detection efficiency. Furthermore, by performing non-linear pixel value transformation on the dynamically range compressed medical image to obtain the target image, the contrast between the cone and intervertebral disc is further improved, enhancing limb detection efficiency.

[0082] Please see Figure 8 This is a flowchart illustrating a method for locating an intervertebral disc provided in an embodiment of this specification. Figure 8 As shown, the method described in the embodiments of this specification may include the following steps S301-S310:

[0083] S301, Based on the vertebral region, identify the adjacent vertebrae in the target image;

[0084] In one embodiment, based on the vertebral regions identified from the target image, one or more groups of adjacent vertebrae contained in the target image are identified. In a feasible implementation, the vertebrae can be labeled according to their positions within the vertebral region, thereby determining the arrangement order of the vertebrae in the vertebral region, and adjacent vertebrae are identified in the target image according to the arrangement order. Specifically, the outer contour of each vertebra can be obtained through a contour recognition model, the center point of the outer contour of each vertebra can be calculated, and the vertebrae can be sorted according to the ordinate of the center point to determine the order of the vertebrae.

[0085] S302, the vertebral body region of the adjacent vertebral body is identified as the target vertebral body region;

[0086] Specifically, the vertebral regions where adjacent vertebrae are located are identified within the target vertebral region, and these regions are then identified as the target vertebral region.

[0087] S303, perform vertex identification on the target vertebral region to obtain the vertices of each vertebra in the adjacent vertebral bodies;

[0088] Specifically, vertex identification is performed on the identified target vertebral region to obtain the vertices of each vertebra in adjacent vertebrae. Adjacent vertebrae include two vertebrae connected by an intervertebral disc. It should be noted that the number of vertebral vertices can vary depending on the target location; generally, four vertices are identified for each vertebra. Commonly used algorithms include Harris corner detection, Shi-Tomasi corner detection, and FAST corner detection.

[0089] S304, Based on the coordinates of the vertex, confirm the adjacent vertebral vertebrae vertebrae;

[0090] Specifically, based on the coordinates of the vertices of each vertebra in an adjacent vertebral body, the vertices of interest in the adjacent vertebral body are identified. The vertices of the adjacent vertebral bodies are the vertices on the adjacent sides of the vertices in the adjacent vertebral bodies. For example, the adjacent vertices may include a first vertebral body and a second vertebral body. Then, the distance between the vertices of the first vertebral body and the vertices of the second vertebral body can be determined to obtain the adjacent vertices of the first vertebral body and the second vertebral body that are close to each other.

[0091] In one feasible implementation, the target vertebral region includes the first vertebral region corresponding to the first vertebra and the second vertebral region corresponding to the second vertebra. Adjacent vertebral vertices include the first vertebral vertices of the first vertebra and the second vertebral vertices of the second vertebra. Determining the adjacent vertebral vertices based on their coordinates may include:

[0092] S3041, based on the ordinate of the vertex of the first vertices and the ordinate of the vertex of the second vertices, sort the vertices of the first vertices and the vertices of the second vertices to obtain the first vertices of the first vertices that are closest to the second vertices, and the second vertices of the second vertices that are closest to the first vertices.

[0093] Specifically, the relative positions of the first and second vertebrae can be determined first based on the ordinates of the vertices of the first and second vertebrae. For example, if the sum of the ordinates of the vertices of the first vertebrae is less than the sum of the ordinates of the vertices of the second vertebrae, then the first vertebra is located above the second vertebra. Then, based on the ordinates of the vertices of the first vertebrae, the two vertices with the largest ordinates are identified as the vertices of the first vertebrae closest to the second vertebra. Similarly, the two vertices with the smallest ordinates in the second vertebrae are identified as the vertices of the second vertebra closest to the first vertebra. Please refer to [link to relevant documentation]. Figure 9 , Figure 9This is an example schematic diagram of an intervertebral disc localization method provided in the embodiments of this specification, and the horizontal and vertical coordinates of the vertex can be determined using this coordinate system.

[0094] In another feasible implementation, the vertices of individual vertebrae can be marked first, and then the vertices of adjacent vertebrae can be identified. Taking four vertices of a single vertices as an example, the four vertices are sorted by their horizontal coordinates. The two vertices with smaller horizontal coordinates are the left points, and the two with larger horizontal coordinates are the right points. Then, the left points are sorted by their vertical coordinates, with the smaller vertical coordinate being the upper left point and the larger vertical coordinate being the lower left point. Similarly, among the right points, the smaller vertical coordinate is the upper right point and the larger vertical coordinate is the lower right point. Then, the upper left, upper right, lower left, and lower right points are labeled. If the first vertices are the vertices preceding the second vertices, then the lower left and lower right points of the first vertices are close to the upper left and upper right points of the second vertices. The lower left and lower right points of the first vertices are the vertices of the first vertices, and the upper left and upper right points of the second vertices are the vertices of the second vertices.

[0095] S305, in the first vertices region, confirm the non-zero pixel path between the vertices of the first vertices and obtain the adjacent edge pixels of the first vertices;

[0096] Specifically, the Bresenham algorithm can be used to search for non-zero pixel paths between the two vertices of the first cone. That is, the edge of the first cone region between the vertices of the first cone is identified, and the edge pixels constituting that edge are obtained.

[0097] S306, in the second vertices region, confirm the non-zero pixel path between the vertices of the second vertices and obtain the adjacent edge pixels of the second vertices;

[0098] Similarly, the Bresenham algorithm is used to search for non-zero pixel paths between the vertices of the second vertebra to obtain the adjacent edge pixels of the second vertebra.

[0099] S307, the adjacent edge pixels of the first vertebra and the adjacent edge pixels of the second vertebra are used as the adjacent edge pixels of the adjacent vertebrae;

[0100] S308, confirm the centroid of the adjacent edge pixel, and use the centroid as the anchor point of the intervertebral disc positioning line of the adjacent vertebral body;

[0101] Specifically, the horizontal and vertical coordinates of the centroid are the average horizontal and vertical coordinates of the edge pixels, respectively. By statistically analyzing the horizontal and vertical coordinates of the adjacent edge pixels, the corresponding centroid can be identified. This centroid can be used as the anchor point for the positioning line of the intervertebral disc of the adjacent vertebral body to confirm the positioning line of the intervertebral disc.

[0102] S309, confirm the main vector direction of the adjacent edge pixels;

[0103] Specifically, the principal vector direction is the direction of the principal vector of the adjacent edge pixel set in pixel coordinates. The principal vector represents the main direction of change of the vertebral body boundary. Therefore, the direction of the intervertebral disc positioning line can be determined by using the principal vector directions of adjacent vertebrae, and the principal component analysis (PCA) algorithm can be used to determine the principal vector direction. For example, the average of the principal vectors of two adjacent vertebrae can be taken as the direction of the intervertebral disc positioning line; another method is to directly calculate the principal vector direction of the edge pixel set.

[0104] S310, based on the intervertebral disc positioning line anchor point and the main vector direction, confirm the intervertebral disc positioning line of the adjacent vertebral body, and confirm the intervertebral disc positioning line as the intervertebral disc positioning result.

[0105] Specifically, a straight line is determined by the anchor point of the intervertebral disc positioning line and the direction of the principal vector, which serves as the intervertebral disc positioning line.

[0106] Please see Figure 10 , Figure 10 This is a schematic diagram illustrating an example of an intervertebral disc localization method provided in the embodiments of this specification, such as... Figure 10 As shown, a given vertebral region is obtained, and the target vertebral region where adjacent vertebrae are located is identified within the vertebral region. The edge pixels of the adjacent edges are extracted based on the vertebral vertebrae's vertebral vertebrae, and the centroid of the adjacent edge pixels is calculated as the anchor point of the intervertebral disc localization line. The horizontal and vertical coordinates of the centroid are the mean horizontal and vertical coordinates of the edge pixels, respectively. At the same time, the PCA algorithm is used to calculate the principal vector direction of the edge pixels. The intervertebral disc localization line is determined based on the principal vector direction and the anchor point coordinates. Using the above method can improve the robustness of intervertebral disc localization between L5 and S1, which are prone to compression, and has good results in general intervertebral disc localization.

[0107] In the embodiments of this specification, adjacent vertebrae in the target image are identified through the vertebral body region. The vertebral body regions of the adjacent vertebrae are identified as the target vertebral body regions. Vertex recognition is performed on the target vertebral body regions to obtain the vertices of each vertebra in the adjacent vertebrae. The vertices of the adjacent vertebrae are identified based on the coordinates of the vertices. The non-zero pixel paths between the vertices of the first vertebra are identified in the first vertebra region corresponding to the first vertebra of the adjacent vertebra, and the adjacent edge pixels of the first vertebra are obtained. The non-zero pixel paths between the vertices of the second vertebra are identified in the second vertebra region corresponding to the second vertebra of the adjacent vertebra, and the adjacent edge pixels of the second vertebra are obtained. The adjacent edge pixels of the first vertebra and the adjacent edge pixels of the second vertebra are used as the adjacent edge pixels of the adjacent vertebrae. The centroid of the adjacent edge pixels is identified and used as the anchor point of the intervertebral disc positioning line of the adjacent vertebrae. The principal vector direction of the adjacent edge pixels is identified. The intervertebral disc positioning line of the adjacent vertebrae is identified based on the anchor point and the principal vector direction. The intervertebral disc positioning line is identified as the intervertebral disc positioning result. After detecting the cone using the target recognition model, the principal vector and centroid are determined based on the edge pixels of adjacent cones. Both the principal vector and centroid are robust to edge pixel values ​​where errors occur, which can improve the robustness of intervertebral disc localization and reduce intervertebral disc localization errors.

[0108] The following will be combined with the appendix Figure 11 This document provides a detailed description of the intervertebral disc positioning device provided in the embodiments of this specification. It should be noted that the appendix... Figure 11 The intervertebral disc positioning device in this manual is used to perform the functions described herein. Figures 2-10 The methods shown in the embodiments are illustrated for ease of explanation, showing only the parts related to the embodiments of this specification. For specific technical details not disclosed, please refer to this specification. Figures 2-10 The example shown.

[0109] Please see Figure 11 This illustration shows a schematic diagram of the structure of an intervertebral disc positioning device provided in an exemplary embodiment of this application. The intervertebral disc positioning device can be implemented as all or part of a device through software, hardware, or a combination of both. The device 1 includes an acquisition module 11, a preprocessing module 12, a model detection module 13, an adjacent vertebral body positioning module 14, a vertebral body edge confirmation module 15, and a positioning result confirmation module 16.

[0110] The acquisition module 11 is used to acquire medical images of the target area; the target area is any part of the spine.

[0111] The preprocessing module 12 is used to preprocess the medical image to obtain the target image;

[0112] The model detection module 13 is used to input the target image into a trained target detection model to obtain the cone region in the target image;

[0113] Adjacent vertebral body positioning module 14 is used to identify the target vertebral body region and adjacent vertebral body apex in the target image based on the vertebral body region;

[0114] The vertebral body edge confirmation module 15 is used to confirm the adjacent edge pixels of the adjacent vertebrae based on the target vertebral body region and the adjacent vertebral body vertices.

[0115] The positioning result confirmation module 16 is used to confirm the intervertebral disc positioning result of the adjacent vertebrae based on the adjacent edge pixels.

[0116] Optionally, the preprocessing module 12 is specifically used to obtain the dynamic range compression parameters corresponding to the target part, the dynamic range compression parameters including the upper limit of the dynamic range and the lower limit of the dynamic range;

[0117] Pixels in the medical image that are above the upper limit of the dynamic range are adjusted to the upper limit of the dynamic range, and pixels in the medical image that are below the lower limit of the dynamic range are adjusted to the lower limit of the dynamic range to obtain the target image.

[0118] Optionally, the preprocessing module 12 is specifically used to adjust pixels in the medical image that are above the upper limit of the dynamic range to the upper limit of the dynamic range, and to adjust pixels in the medical image that are below the lower limit of the dynamic range to the lower limit of the dynamic range, so as to obtain a medical image after dynamic range compression processing.

[0119] The medical image after dynamic range compression is subjected to nonlinear pixel value conversion to obtain the target image.

[0120] Optionally, the preprocessing module 12 is further configured to acquire a set of historical images of the target area and adjust the dynamic range of each historical image in the set of historical images to the optimal observation range;

[0121] The mean of the optimal observation range is confirmed, the upper limit of the mean is confirmed as the upper limit of the dynamic range of the target part, and the lower limit of the mean is confirmed as the lower limit of the dynamic range of the target part.

[0122] The upper limit of the dynamic range and the lower limit of the dynamic range are determined as the dynamic range compression parameters of the target part.

[0123] Optionally, the positioning result confirmation module 16 is specifically used to confirm the centroid of the adjacent edge pixel and use the centroid as the anchor point of the intervertebral disc positioning line of the adjacent vertebral body;

[0124] Confirm the principal vector direction of the adjacent edge pixels;

[0125] Based on the intervertebral disc positioning line anchor point and the principal vector direction, the intervertebral disc positioning line of the adjacent vertebral body is confirmed, and the intervertebral disc positioning line is confirmed as the intervertebral disc positioning result.

[0126] Optionally, the adjacent vertebral body positioning module 14 is specifically used to identify adjacent vertebrae in the target image based on the vertebral body region;

[0127] The vertebral regions of the adjacent vertebrae are identified as the target vertebral regions;

[0128] Vertex identification is performed on the target vertebral region to obtain the vertices of each vertebra in the adjacent vertebral bodies;

[0129] The vertices of adjacent vertebrae are identified based on the coordinates of the vertices.

[0130] Optionally, the target vertebral region includes the first vertebral region corresponding to the first vertebra and the second vertebral region corresponding to the second vertebra. The adjacent vertebral vertebrae include the first vertebral vertebrae of the first vertebra and the second vertebral vertebrae of the second vertebra. The adjacent vertebral positioning module 14 is specifically used to sort the vertices of the first vertebra and the second vertebra based on the ordinate of the vertices of the first vertebra and the ordinate of the vertices of the second vertebra, so as to obtain the first vertebral vertices of the first vertebra that are closest to the second vertebra, and the second vertebral vertices of the second vertebra that are closest to the first vertebra.

[0131] Optionally, the target vertebral region includes the first vertebral region corresponding to the first vertebra and the second vertebral region corresponding to the second vertebra. The adjacent vertebral vertebral vertices include the first vertebral vertices of the first vertebra and the second vertebral vertices of the second vertebra. The vertebral edge confirmation module 15 is specifically used to confirm the non-zero pixel path between the first vertebral vertices in the first vertebral region to obtain the adjacent edge pixels of the first vertebra.

[0132] In the second vertices region, identify the non-zero pixel paths between the vertices of the second vertices to obtain the adjacent edge pixels of the second vertices.

[0133] The adjacent edge pixels of the first vertebra and the adjacent edge pixels of the second vertebra are used as the adjacent edge pixels of the adjacent vertebrae.

[0134] It should be noted that the disc positioning device provided in the above embodiments is only illustrated by the division of the functional modules described above when performing the disc positioning method. In practical applications, the functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the disc positioning device and the disc positioning method embodiments provided in the above embodiments belong to the same concept, and the implementation process is detailed in the method embodiments, which will not be repeated here.

[0135] The embodiment numbers in this specification are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments. In some cases, the actions or steps described in the claims can be performed in a different order than that shown in the embodiments and still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0136] This specification also provides a storage medium storing a computer program, which, when executed by a processor, implements the above-described functionality. Figures 2-10 The method described in the illustrated embodiment can be found in the following document for a detailed execution process. Figures 2-10 The specific details of the illustrated embodiments will not be elaborated here.

[0137] Please refer to Figure 12 This diagram illustrates the structure of an electronic device provided in an exemplary embodiment of this specification. The electronic device in this specification may include one or more components such as a processor 110, a memory 120, an input device 130, an output device 140, and a bus 150. The processor 110, memory 120, input device 130, and output device 140 may be connected via the bus 150.

[0138] Processor 110 may include one or more processing cores. Processor 110 connects to various parts of the electronic device using various interfaces and lines, and executes various functions of terminal 100 and processes data by running or executing instructions, programs, code sets, or instruction sets stored in memory 120, and by calling data stored in memory 120. Optionally, processor 110 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). Processor 110 may integrate one or more of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user page, and applications; the GPU is responsible for rendering and drawing the displayed content; and the modem handles wireless communication. It is understood that the modem may also not be integrated into processor 110 and may be implemented separately using a communication chip.

[0139] The memory 120 may include random access memory (RAM) or read-only memory (ROM). Optionally, the memory 120 may include non-transitory computer-readable storage medium. The memory 120 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 120 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for implementing at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the various method embodiments described above, etc. The operating system may be the Android system, including systems deeply developed based on the Android system, the iOS system developed by Apple Inc., including systems deeply developed based on the iOS system, or other systems.

[0140] The memory 120 can be divided into operating system space and user space. The operating system runs in the operating system space, while native and third-party applications run in user space. To ensure that different third-party applications can achieve good running performance, the operating system allocates corresponding system resources for each application. However, different application scenarios within the same third-party application have different requirements for system resources. For example, in local resource loading scenarios, third-party applications have high requirements for disk read speed; in animation rendering scenarios, third-party applications have high requirements for GPU performance. Since the operating system and third-party applications are independent of each other, the operating system often cannot promptly perceive the current application scenario of a third-party application, resulting in the operating system's inability to adapt system resources accordingly.

[0141] In order for the operating system to distinguish the specific application scenarios of third-party applications, it is necessary to establish data communication between the third-party applications and the operating system. This would allow the operating system to obtain the current scenario information of the third-party applications at any time, and then perform targeted system resource adaptation based on the current scenario.

[0142] The input device 130 is used to receive input instructions or data, and includes, but is not limited to, a keyboard, mouse, camera, microphone, or touch device. The output device 140 is used to output instructions or data, and includes, but is not limited to, a display device and a speaker. In one example, the input device 130 and the output device 140 can be combined, and the input device 130 and the output device 140 can be a touch display screen.

[0143] The touch display screen can be designed as a full-screen, curved screen, or irregularly shaped screen. It can also be designed as a combination of a full-screen and a curved screen, or a combination of an irregularly shaped screen and a curved screen; however, this specification does not limit the specific design of the embodiments.

[0144] In addition, those skilled in the art will understand that the structure of the electronic device shown in the above figures does not constitute a limitation on the electronic device. The electronic device may include more or fewer components than shown, or combine certain components, or have different component arrangements. For example, the electronic device may also include radio frequency circuits, input units, sensors, audio circuits, WiFi modules, power supplies, Bluetooth modules, etc., which will not be described in detail here.

[0145] exist Figure 12 In the illustrated electronic device, the processor 110 can be used to call a computer program stored in the memory 120 and specifically perform the following operations:

[0146] Acquire medical images of the target area; the target area is any part of the spine;

[0147] The medical images are preprocessed to obtain the target image;

[0148] The target image is input into a trained target detection model to obtain the cone region in the target image;

[0149] Based on the vertebral region, the target vertebral region and the adjacent vertebral apex in the target image are identified;

[0150] Based on the target vertebral region and the adjacent vertebral vertices, the adjacent edge pixels of the adjacent vertebral bodies are confirmed;

[0151] The intervertebral disc localization result of the adjacent vertebrae is confirmed based on the adjacent edge pixels.

[0152] In one embodiment, when the processor 110 performs preprocessing on the medical image to obtain the target image, it specifically performs the following operations:

[0153] Obtain the dynamic range compression parameters corresponding to the target part, wherein the dynamic range compression parameters include an upper limit of dynamic range and a lower limit of dynamic range;

[0154] Pixels in the medical image that are above the upper limit of the dynamic range are adjusted to the upper limit of the dynamic range, and pixels in the medical image that are below the lower limit of the dynamic range are adjusted to the lower limit of the dynamic range to obtain the target image.

[0155] In one embodiment, when the processor 110 adjusts pixels in the medical image that are above the upper limit of the dynamic range to the upper limit of the dynamic range, and adjusts pixels in the medical image that are below the lower limit of the dynamic range to the lower limit of the dynamic range, in order to obtain a target image, it specifically performs the following operations:

[0156] Pixels in the medical image that are above the upper limit of the dynamic range are adjusted to the upper limit of the dynamic range, and pixels in the medical image that are below the lower limit of the dynamic range are adjusted to the lower limit of the dynamic range, so as to obtain a medical image after dynamic range compression.

[0157] The medical image after dynamic range compression is subjected to nonlinear pixel value conversion to obtain the target image.

[0158] In one embodiment, before acquiring the dynamic range compression parameters corresponding to the target region, the processor 110 also performs the following operations:

[0159] Obtain a set of historical images of the target area, and adjust the dynamic range of each historical image in the set to the optimal observation range;

[0160] The mean of the optimal observation range is confirmed, the upper limit of the mean is confirmed as the upper limit of the dynamic range of the target part, and the lower limit of the mean is confirmed as the lower limit of the dynamic range of the target part.

[0161] The upper limit of the dynamic range and the lower limit of the dynamic range are determined as the dynamic range compression parameters of the target part.

[0162] In one embodiment, when the processor 110 executes the intervertebral disc localization result based on the adjacent edge pixels to confirm the intervertebral disc location of the adjacent vertebral body, it specifically performs the following operations:

[0163] Identify the centroid of the adjacent edge pixels and use the centroid as the anchor point for the intervertebral disc positioning line of the adjacent vertebral body;

[0164] Confirm the principal vector direction of the adjacent edge pixels;

[0165] Based on the intervertebral disc positioning line anchor point and the principal vector direction, the intervertebral disc positioning line of the adjacent vertebral body is confirmed, and the intervertebral disc positioning line is confirmed as the intervertebral disc positioning result.

[0166] In one embodiment, when the processor 110 performs the operation of identifying the target vertebral region and adjacent vertebral vertebrae in the target image based on the vertebral region, it specifically performs the following operations:

[0167] Based on the vertebral region, adjacent vertebrae in the target image are identified;

[0168] The vertebral regions of the adjacent vertebrae are identified as the target vertebral regions;

[0169] Vertex identification is performed on the target vertebral region to obtain the vertices of each vertebra in the adjacent vertebral bodies;

[0170] The vertices of adjacent vertebrae are identified based on the coordinates of the vertices.

[0171] In one embodiment, the target vertebral region includes a first vertebral region corresponding to a first vertebra and a second vertebral region corresponding to a second vertebra. The adjacent vertebral vertices include the first vertebral vertices of the first vertebra and the second vertebral vertices of the second vertebra. When the processor 110 performs the operation of confirming the adjacent vertebral vertices based on the coordinates of the vertices, it specifically performs the following operations:

[0172] Based on the ordinate of the first vertices and the ordinate of the second vertices, the vertices of the first vertices and the second vertices are sorted to obtain the first vertices of the first vertices that are closest to the second vertices, and the second vertices of the second vertices that are closest to the first vertices.

[0173] In one embodiment, the target vertebral region includes a first vertebral region corresponding to a first vertebra and a second vertebral region corresponding to a second vertebra. The adjacent vertebral vertices include the first vertebral vertices of the first vertebra and the second vertebral vertices of the second vertebra. When the processor 110 determines the adjacent edge pixels of the adjacent vertebrae based on the target vertebral region and the adjacent vertices, it specifically performs the following operations:

[0174] In the first vertices region, identify the non-zero pixel paths between the vertices of the first vertices to obtain the adjacent edge pixels of the first vertices.

[0175] In the second vertices region, identify the non-zero pixel paths between the vertices of the second vertices to obtain the adjacent edge pixels of the second vertices.

[0176] The adjacent edge pixels of the first vertebra and the adjacent edge pixels of the second vertebra are used as the adjacent edge pixels of the adjacent vertebrae.

[0177] In the embodiments of this specification, medical images of the target area are acquired, preprocessed to obtain a target image, and input into a trained target detection model to obtain the vertebral region in the target image. Based on the vertebral region, the target vertebral region and adjacent vertebral vertebrae in the target image are identified. Based on the target vertebral region and adjacent vertebral vertebrae vertebrae vertebrae vertebrae, the adjacency edge pixels of adjacent vertebrae are identified. Based on the adjacency edge pixels, the intervertebral disc localization result of adjacent vertebrae is confirmed. The target detection model efficiently locates the vertebral region, and the robustness of the vertebral direction is characterized by the adjacency edges of adjacent vertebrae, thereby improving the robustness and accuracy of intervertebral disc localization.

[0178] Furthermore, by obtaining the dynamic range compression parameters corresponding to the target area, and based on the upper and lower limits of the dynamic range parameters, pixels in the medical image above the upper limit are adjusted to the upper limit, and pixels below the lower limit are adjusted to the lower limit to obtain the target image. Dynamic range compression improves image contrast and cone detection efficiency. In addition, by performing non-linear pixel value transformation on the dynamically range compressed medical image to obtain the target image, the contrast between the cone and intervertebral disc is further improved, thus increasing limb detection efficiency.

[0179] Furthermore, by identifying the vertebral regions, adjacent vertebrae in the target image are confirmed, and the vertebral regions of the adjacent vertebrae are identified as the target vertebral regions. Vertex recognition is performed on the target vertebral regions to obtain the vertices of each vertebra in the adjacent vertebrae. Based on the coordinates of the vertices, the vertices of the adjacent vertebrae are confirmed. In the first vertebra region corresponding to the first vertebra in the adjacent vertebrae, the non-zero pixel path between the vertices of the first vertebra is confirmed, and the adjacent edge pixels of the first vertebra are obtained. In the second vertebra region corresponding to the second vertebra in the adjacent vertebrae, the non-zero pixel path between the vertices of the second vertebra is confirmed, and the adjacent edge pixels of the second vertebra are obtained. The adjacent edge pixels of the first vertebra and the adjacent edge pixels of the second vertebra are used as the adjacent edge pixels of the adjacent vertebrae. The centroid of the adjacent edge pixels is confirmed, and the centroid is used as the anchor point of the intervertebral disc positioning line of the adjacent vertebrae. The principal vector direction of the adjacent edge pixels is confirmed. Based on the anchor point and principal vector direction of the intervertebral disc positioning line, the intervertebral disc positioning line of the adjacent vertebrae is confirmed, and the intervertebral disc positioning line is confirmed as the intervertebral disc positioning result. After detecting the cone using the target recognition model, the principal vector and centroid are determined based on the edge pixels of adjacent cones. Both the principal vector and centroid are robust to edge pixel values ​​where errors occur, which can improve the robustness of intervertebral disc localization and reduce intervertebral disc localization errors.

[0180] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc. The above-disclosed embodiments are merely preferred embodiments of this specification and should not be construed as limiting the scope of this specification. Therefore, equivalent variations made according to the claims of this specification are still within the scope of this specification.

Claims

1. A method for locating an intervertebral disc, characterized in that, The method includes: Acquire medical images of the target area; the target area is any part of the spine; The medical images are preprocessed to obtain the target image; The target image is input into a trained target detection model to obtain the cone region in the target image; Based on the vertebral region, the target vertebral region and the adjacent vertebral apex in the target image are identified; Based on the target vertebral region and the adjacent vertebral vertebrae, the adjacent edge pixels of the adjacent vertebrae are confirmed; The intervertebral disc localization result of the adjacent vertebrae is confirmed based on the adjacent edge pixels; The method of confirming the intervertebral disc localization result of the adjacent vertebrae based on the adjacent edge pixels includes: Identify the centroid of the adjacent edge pixels and use the centroid as the anchor point for the intervertebral disc positioning line of the adjacent vertebral body; Confirm the principal vector direction of the adjacent edge pixels; Based on the intervertebral disc positioning line anchor point and the principal vector direction, the intervertebral disc positioning line of the adjacent vertebral body is confirmed, and the intervertebral disc positioning line is confirmed as the intervertebral disc positioning result.

2. The method as described in claim 1, characterized in that, The preprocessing of the medical image to obtain the target image includes: Obtain the dynamic range compression parameters corresponding to the target part, wherein the dynamic range compression parameters include an upper limit of dynamic range and a lower limit of dynamic range; Pixels in the medical image that are above the upper limit of the dynamic range are adjusted to the upper limit of the dynamic range, and pixels in the medical image that are below the lower limit of the dynamic range are adjusted to the lower limit of the dynamic range to obtain the target image.

3. The method as described in claim 2, characterized in that, The step of adjusting pixels in the medical image that are above the upper limit of the dynamic range to the upper limit of the dynamic range, and adjusting pixels in the medical image that are below the lower limit of the dynamic range to the lower limit of the dynamic range, to obtain the target image, includes: Pixels in the medical image that are above the upper limit of the dynamic range are adjusted to the upper limit of the dynamic range, and pixels in the medical image that are below the lower limit of the dynamic range are adjusted to the lower limit of the dynamic range, so as to obtain a medical image after dynamic range compression. The medical image after dynamic range compression is subjected to nonlinear pixel value conversion to obtain the target image.

4. The method as described in claim 2, characterized in that, Before obtaining the dynamic range compression parameters corresponding to the target part, the method further includes: Obtain a set of historical images of the target area, and adjust the dynamic range of each historical image in the set to the optimal observation range; The mean of the optimal observation range is confirmed, the upper limit of the mean is confirmed as the upper limit of the dynamic range of the target part, and the lower limit of the mean is confirmed as the lower limit of the dynamic range of the target part. The upper limit of the dynamic range and the lower limit of the dynamic range are determined as the dynamic range compression parameters of the target part.

5. The method as described in claim 1, characterized in that, The step of identifying the target vertebral region and adjacent vertebral vertebrae in the target image based on the vertebral region includes: Based on the vertebral region, adjacent vertebrae in the target image are identified; The vertebral regions of the adjacent vertebrae are identified as the target vertebral regions; Vertex identification is performed on the target vertebral region to obtain the vertices of each vertebra in the adjacent vertebral bodies; The adjacent vertices of the adjacent vertices are identified based on the coordinates of the vertices.

6. The method as described in claim 5, characterized in that, The target vertebral region includes the first vertebral region corresponding to the first vertebra and the second vertebral region corresponding to the second vertebra; the adjacent vertebral apex includes the first vertebral apex of the first vertebra and the second vertebral apex of the second vertebra; The process of identifying adjacent vertebral vertices based on the coordinates of the vertices includes: Based on the ordinate of the first vertices and the ordinate of the second vertices, the vertices of the first vertices and the second vertices are sorted to obtain the first vertices of the first vertices that are closest to the second vertices, and the second vertices of the second vertices that are closest to the first vertices.

7. The method as described in any one of claims 1 or 6, characterized in that, The target vertebral region includes the first vertebral region corresponding to the first vertebra and the second vertebral region corresponding to the second vertebra; the adjacent vertebral apex includes the first vertebral apex of the first vertebra and the second vertebral apex of the second vertebra; The step of determining the adjacent edge pixels of the adjacent vertebrae based on the target vertebral region and the adjacent vertebral vertebrae includes: In the first vertices region, identify the non-zero pixel paths between the vertices of the first vertices to obtain the adjacent edge pixels of the first vertices. In the second vertices region, identify the non-zero pixel paths between the vertices of the second vertices to obtain the adjacent edge pixels of the second vertices. The adjacent edge pixels of the first vertebra and the adjacent edge pixels of the second vertebra are used as the adjacent edge pixels of the adjacent vertebrae.

8. A disc positioning device, characterized in that, The device includes: The acquisition module is used to acquire medical images of the target area; the target area is any part of the spine. A preprocessing module is used to preprocess the medical image to obtain the target image; The model detection module is used to input the target image into a trained target detection model to obtain the cone region in the target image; The adjacent vertebral body localization module is used to identify the target vertebral body region and the adjacent vertebral body apex in the target image based on the vertebral body region. The vertebral body edge confirmation module is used to confirm the adjacent edge pixels of the adjacent vertebrae based on the target vertebral body region and the adjacent vertebral body vertebrae. The positioning result confirmation module is used to confirm the intervertebral disc positioning result of the adjacent vertebrae based on the adjacent edge pixels. The positioning result confirmation module is specifically used to confirm the centroid of the adjacent edge pixels and use the centroid as the anchor point of the intervertebral disc positioning line of the adjacent vertebral body. Confirm the principal vector direction of the adjacent edge pixels; Based on the intervertebral disc positioning line anchor point and the principal vector direction, the intervertebral disc positioning line of the adjacent vertebral body is confirmed, and the intervertebral disc positioning line is confirmed as the intervertebral disc positioning result.

9. An electronic device, characterized in that, include: Processor and memory; The memory stores a computer program adapted to be loaded by the processor and to execute the steps of the method as described in any one of claims 1 to 7.

10. A storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 7.

Images