A long bone fracture reduction method and system based on image processing

Abstract

The present invention relates to the technical field of fracture reduction assistance, and particularly relates to a long bone fracture reduction method and system based on image processing. The present invention obtains the upper fracture edge of the upper broken bone region and the lower fracture edge of the lower broken bone region in a long bone fracture image, selects an upper target base point from the upper fracture edge according to the position and sharpness of the upper edge pixel points of the upper fracture edge; selects a lower target base point on the lower fracture edge according to the position of the upper target base point on the upper fracture edge and the relative size of the widths of the upper broken bone region and the lower broken bone region; determines the reduction path of the long bone fracture according to the relative positions of the upper target base point and the lower target base point, and the direction trend of the upper fracture edge; moves the upper broken bone region along the reduction path to assist in the reduction of the long bone fracture. The present invention can avoid collisions during the fracture reduction process and the broken bone has the shortest movement path, improving the reduction effect on long bone fractures.

Description

Technical Field

[0001] The present invention relates to the technical field of fracture reduction assistance, and in particular to a long bone fracture reduction method and system based on image processing. Background Art

[0002] Long bones are a type of bone that is long and columnar, such as the femur and tibia. Fracture reduction methods include manual reduction, traction reduction, and surgical incision reduction. Traditional fracture reduction surgeries, such as tibia fracture reduction, usually use intraoperative X-ray images to understand the state of the broken bone during the reduction process, and then multiple doctors pull the patient's muscles to achieve fracture reduction. Existing methods usually preset the path of fracture reduction based on laboratory theoretical conditions, which results in an excessive range of movement, causing iatrogenic soft tissue trauma to the patient, resulting in poor reduction of long bone fractures. Summary of the invention

[0003] In order to solve the technical problem that the preset fracture reduction path based on the theoretical situation in the laboratory has too large a moving range, resulting in poor reduction effect of long bone fractures, the purpose of the present invention is to provide a long bone fracture reduction method and system based on image processing, and the technical scheme adopted is as follows:

[0004] In a first aspect, an embodiment of the present invention provides a method for reducing a long bone fracture based on image processing, the method comprising:

[0005] Obtain long bone fracture images of the patient's fracture site;

[0006] Performing regional segmentation on the long bone fracture image to obtain the upper fracture region and the lower fracture region; obtaining the upper fracture edge and the lower fracture edge;

[0007] According to the position and sharpness of the pixel point on the upper edge of the upper fracture edge, an upper target base point on the upper fracture edge is selected; according to the position of the upper target base point on the upper fracture edge and the relative size of the width of the upper fracture area and the lower fracture area, a lower target base point is selected from the lower fracture edge;

[0008] According to the relative positions of the upper target base point and the lower target base point, and the direction of the upper fracture edge, the reduction path of the long bone fracture is determined; the upper fracture bone area is moved along the reduction path to assist in the reduction of the long bone fracture.

[0009] Further, obtaining the upper fracture edge and the lower fracture edge includes:

[0010] Obtain corner points in the long bone fracture image; record the upper fractured bone region and the lower fractured bone region as fractured bone regions, and divide each fractured bone region into different regional blocks;

[0011] For each regional block, a pair of corner points is formed by any two corner points within the regional block, and the average value of the distances between the two corner points in all pairs of corner points of the regional block is taken as the overall corner distance value of the regional block; according to the total number of corner points within the regional block and the overall corner distance value, the fracture index of the regional block is obtained; the total number of corner points within the regional block and the fracture index are in a positive correlation, and the overall corner distance value and the fracture index are in a negative correlation;

[0012] For the regional blocks of each fractured bone region, the regional blocks corresponding to the fracture indices greater than the preset fracture threshold are denoted as fracture region blocks; the edge obtained by performing edge detection on the region formed by all the fracture region blocks of the fractured bone region is taken as the fracture edge;

[0013] The fracture edge of the upper fractured bone region is respectively denoted as the upper fracture edge, and the fracture edge of the lower fractured bone region is denoted as the lower fracture edge.

[0014] Further, the selection of the upper target base point on the upper fracture edge includes:

[0015] Arbitrarily select an edge pixel point on the upper fracture edge as the example point. On the upper fracture edge, the adjacent previous edge pixel point and the adjacent subsequent edge pixel point of the example point are denoted as the adjacent analysis points of the example point;

[0016] The line segments connecting the example point with each of its adjacent analysis points are denoted as the analysis edges of the example point; the included angle between the two analysis edges of the example point is taken as the sharpness index of the example point;

[0017] According to the ordinate of the pixel coordinates of each edge pixel point on the upper fracture edge and the sharpness index, the base point judgment index corresponding to the corresponding edge pixel point is obtained; the ordinate and the base point judgment index are in a positive correlation, and the sharpness index and the base point judgment index are in a negative correlation;

[0018] The edge pixel point corresponding to the largest base point judgment index on the upper fracture edge is denoted as the upper target base point.

[0019] Further, the selection of the lower target base point from the lower fracture edge includes:

[0020] For each fractured bone region, morphological operations are performed on the fractured bone region to obtain the skeleton line of the fractured bone region; perpendicular lines are respectively drawn for each pixel point on the skeleton line, and the line segments obtained by connecting the two intersection points of the perpendicular lines of each pixel point on the skeleton line with the edge of the fractured bone region are denoted as the width line segments of each pixel point on the skeleton line;

[0021] Calculate the mean of the absolute values of the differences between the lengths of the width line segments of each pixel point on the bone skeleton line and the width line segments of its adjacent analysis points, and use it as the width difference value of each pixel point on the bone skeleton line; mark the pixel points on the bone skeleton line with width difference values less than the preset difference threshold as the width judgment points of the broken bone area;

[0022] Mark the width line segment of the width judgment point corresponding to the maximum value of the ordinate of the pixel coordinates of the width judgment points in the upper broken bone area as the upper long bone width line segment; mark the width line segment of the width judgment point corresponding to the minimum value of the ordinate of the pixel coordinates of the width judgment points in the lower broken bone area as the lower long bone width line segment;

[0023] Obtain the perpendicular point of the upper target base point on the upper long bone width line segment, and mark it as the upper target point; select a lower target point from the lower long bone width line segment, and the distance between the lower target point and the left endpoint of the lower long bone width line segment is equal to the distance between the upper target point and the left endpoint of the upper long bone width line segment;

[0024] Draw a perpendicular line from the upper and lower target points of the lower long bone width line segment, and mark the intersection point of the perpendicular line of the lower target point and the lower fracture edge as the suspected lower target base point; for the edge pixel points within the preset local range of the upper and lower target base points on the lower fracture edge, take the edge pixel point corresponding to the maximum base point judgment index as the lower target base point.

[0025] Further, the determining the reduction path of the long bone fracture according to the relative positions of the upper target base point and the lower target base point, and the direction trend of the upper fracture edge includes:

[0026] Obtain the reduction midpoint according to the relative positions of the upper target base point and the lower target base point, and the direction trend of the upper fracture edge;

[0027] Take the straight line from the upper target base point to the reduction midpoint as the first path;

[0028] Take the straight line from the reduction midpoint to the lower target base point as the second path;

[0029] The reduction path is composed of the first path and the second path.

[0030] Further, the obtaining the reduction midpoint includes:

[0031] If the abscissa of the pixel coordinates of the upper target base point is greater than the abscissa of the pixel coordinates of the lower target base point, then the edge between the left endpoint of the upper fracture edge and the upper target base point is recorded as the direction analysis edge, and the edge between the right endpoint of the lower fracture edge and the lower target base point is recorded as the collision analysis edge; if the abscissa of the pixel coordinates of the upper target base point is less than the abscissa of the pixel coordinates of the target base point, then the edge between the right endpoint of the upper fracture edge and the upper target base point is recorded as the direction analysis edge, and the edge between the left endpoint of the lower fracture edge and the lower target base point is recorded as the collision analysis edge;

[0032] Perform linear fitting on the edge pixel points on the direction analysis edge to obtain a direction line; obtain the length analysis line segment of the lower target base point, and there is an intersection between the length analysis line segment and the collision analysis edge; record the intersection of the direction line and the length analysis line segment as the reset midpoint.

[0033] Further, the region segmentation of the long bone fracture image to obtain the upper fractured bone region and the lower fractured bone region includes:

[0034] Use the maximum inter-class variance method to obtain the segmentation threshold of the long bone fracture image; in the long bone fracture image, record the connected domain composed of pixel points with gray values greater than the segmentation threshold as the fractured bone region; there are two fractured bone regions;

[0035] Obtain the geometric center of the fractured bone region; for the ordinates of the pixel coordinates of the geometric centers of the two fractured bone regions, record the fractured bone region corresponding to the smallest ordinate as the upper fractured bone region, and the fractured bone region corresponding to the largest ordinate as the lower fractured bone region.

[0036] Further, the moving the upper fractured bone region along the reset path includes: moving the upper target base point on the upper fractured bone region along the reset path.

[0037] Further, the algorithm for obtaining the corner points in the long bone fracture image is the Harris corner detection algorithm.

[0038] In a second aspect, another embodiment of the present invention provides a long bone fracture reduction system based on image processing, and the system includes:

[0039] A data acquisition module for acquiring a long bone fracture image of the fracture site of a patient;

[0040] A fracture edge acquisition module for performing region segmentation on the long bone fracture image to obtain an upper fractured bone region and a lower fractured bone region; obtaining an upper fracture edge and a lower fracture edge;

[0041] The target base point screening module is used to select the upper target base point on the upper fracture edge according to the position and sharpness of the upper edge pixel points of the upper fracture edge; and select the lower target base point from the lower fracture edge according to the position of the upper target base point on the upper fracture edge and the relative size of the widths of the upper fracture region and the lower fracture region.

[0042] The auxiliary reduction module is used to determine the reduction path of the long bone fracture according to the relative position between the upper target base point and the lower target base point, and the direction trend of the upper fracture edge; and move the upper fracture region along the reduction path to assist in the reduction of the long bone fracture.

[0043] The present invention has the following beneficial effects:

[0044] In the embodiment of the present invention, the fractured bones of the long bone fracture, namely the upper fracture region and the lower fracture region, are obtained; in order to avoid collision of the fractured bones during the fracture reduction process, it is necessary to ensure that the lowest point of the narrower part of the upper fractured bone does not collide with the lower fractured bone. According to the position and sharpness of the upper edge pixel points of the upper fracture edge, the upper target base point on the upper fracture edge is selected; in order to solve the problem that the cross-sections of the upper and lower fractured bones do not match due to the falling of a small amount of broken bones, based on the position of the upper target base point on the upper fracture edge and combined with the characteristic that the width in the middle of the long bone is relatively close, the lower target base point is screened; the relative position between the upper target base point and the lower target base point determines the direction of the reduction path, and the direction trend of the upper fracture edge determines the length of the direction of the reduction path. By combining the analysis of these two factors, on the premise of ensuring no collision during the fracture reduction process, the length of the movement path is minimized to reduce the damage to other human tissues during the fracture reduction process of the fractured bones and improve the reduction effect of the long bone fracture. Description of the Drawings

[0045] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0046] Figure 1 It is a flowchart of the steps of a method for reducing long bone fractures based on image processing provided by an embodiment of the present invention;

[0047] Figure 2 It is a partial schematic diagram of an image of a long bone fracture provided by an embodiment of the present invention;

[0048] Figure 3 It is a system structure diagram of a system for reducing long bone fractures based on image processing provided by an embodiment of the present invention;

[0049] Figure 4 Schematic diagram of a computer device of a long bone fracture reduction device based on image processing provided by an embodiment of the present invention. Detailed implementation manners

[0050] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation manners, structures, features and effects of a long bone fracture reduction method and system based on image processing according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0051] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs.

[0052] The following specifically describes the specific solutions of a long bone fracture reduction method and system based on image processing provided by the present invention with reference to the accompanying drawings.

[0053] Embodiment 1:

[0054] The present invention proposes a long bone fracture reduction method based on image processing. Please refer to Figure 1 , which shows a flowchart of the steps of a long bone fracture reduction method based on image processing provided by an embodiment of the present invention. The method includes:

[0055] Step S1: Obtain a long bone fracture image of the fracture site of a patient.

[0056] A doctor uses an X-ray scanner to scan the long bone fracture site of a patient from the front to obtain a long bone fracture image; Figure 2 Schematic diagram of a part of a long bone fracture image provided by an embodiment of the present invention. As Figure 2 shown, Figure 2 there are two long bones. The left long bone is fractured, and the right long bone is not fractured. The object of analysis in this solution is the left long bone. The long bone presents a white strip shape and extends downward. The extension of the long bone at the fracture is interrupted, and the fracture site is in a displaced state. The contour of the fracture edge is relatively complex.

[0057] It should be noted that since the X-ray image is a grayscale image, the long bone fracture image is a grayscale image.

[0058] Step S2: Perform region segmentation on the long bone fracture image to obtain an upper fractured bone region and a lower fractured bone region; obtain the upper fracture edge and the lower fracture edge.

[0059] Under normal circumstances, a long bone after fracture will be divided into two or more parts. Since this solution only analyzes a single fracture position of the bone, there are only two fractured bones in the long bone fracture image. As the long bone in the long bone fracture image is nearly vertically distributed, the long bone after fracture is divided into an upper fractured bone and a lower fractured bone, which are denoted as the upper fractured bone region and the lower fractured bone region in sequence. The method for obtaining the two fractured bone regions is as follows:

[0060] Use the Otsu method to obtain the segmentation threshold of the long bone fracture image; in the long bone fracture image, the connected domain composed of pixel points with gray values greater than the segmentation threshold is denoted as the fractured bone region; there are two fractured bone regions; obtain the geometric centers of the fractured bone regions; for the vertical coordinates of the pixel coordinates of the geometric centers of the two fractured bone regions, the fractured bone region corresponding to the smallest vertical coordinate is denoted as the upper fractured bone region, and the fractured bone region corresponding to the largest vertical coordinate is denoted as the lower fractured bone region. Among them, the Otsu method and pixel coordinates are both well-known technologies to those skilled in the art and will not be elaborated here.

[0061] Since the density of the bone is higher than that of the surrounding soft tissues, the gray values of the bone part are larger. Therefore, the connected domain composed of pixel points with gray values greater than the segmentation threshold in the long bone fracture image is denoted as the fractured bone region. The geometric centroid of the fractured bone region represents the approximate position of the fractured bone region; because the dislocation state of the fracture site of the long bone does not change the position of the long bone, and the positive direction of the vertical axis of the image coordinate system is vertically downward, the vertical coordinate of the geometric center of the upper fractured bone region is smaller than that of the lower fractured bone region.

[0062] The fracture edges of the fracture site can help doctors accurately identify the specific position, direction, and shape of the fracture, so as to determine the reduction path of the fracture. Therefore, it is necessary to obtain the upper fracture edge of the upper fractured bone region and the lower fracture edge of the lower fractured bone region. The specific obtaining method is as follows:

[0063] Obtain the corner points in the long bone fracture image; denote the upper fractured bone region and the lower fractured bone region as the fractured bone regions, and divide each fractured bone region into different region blocks; for each region block, form corner point pairs from any two corner points within the region block, and take the average value of the distances between the two corner points in all corner point pairs of the region block as the overall corner point distance value of the region block; according to the total number of corner points within the region block and the overall corner point distance value, obtain the fracture index of the region block; the total number of corner points within the region block and the fracture index are positively correlated, and the overall corner point distance value and the fracture index are negatively correlated; for the region blocks of each fractured bone region, denote the region blocks corresponding to the fracture indices greater than the preset fracture threshold as the fracture region blocks; take the edge obtained by performing edge detection on the region formed by all the fracture region blocks of the fractured bone region as the fracture edge; denote the fracture edge of the upper fractured bone region as the upper fracture edge and the fracture edge of the lower fractured bone region as the lower fracture edge respectively.

[0064] It should be noted that in the embodiment of the present invention, the Harris corner detection algorithm is selected to detect the corners of the long bone fracture image, and the corners in the long bone fracture image are obtained; the broken bone area is divided into regional blocks of uniform size, and the size of the regional block is , and the implementer can set the size of the regional block according to the specific situation. If the total number of corner points in the regional block is 0 or 1, then in this embodiment, the overall corner point distance value of the regional block is directly set to the constant 1. Among them, the Harris corner detection algorithm is a well-known technology to those skilled in the art and will not be elaborated here.

[0065] The fracture surface of the long bone is usually uneven, and the edge contour of the fracture surface is relatively complex, that is, the trend direction of the edge contour changes greatly and the direction changes frequently, and there are many corner points; while the surface of the long bone is smooth, and the number of corner points on both sides of the broken bone area is small. The number of corner points in the regional block at the fracture position of the long bone is large, making the distance between different corner points in the regional block small. Therefore, the total number of corner points in the regional block has a positive correlation with the fracture index, and the overall corner point distance value has a negative correlation with the fracture index. Among them, the distance between different corner points in the process of obtaining the overall corner point distance value refers to the Euclidean distance.

[0066] In the embodiment of the present invention, the ratio of the total number of corner points in each regional block to the overall corner point distance value is normalized to obtain the fracture index of the corresponding regional block. In the embodiment of the present invention, the correlation between the total number of corner points in the regional block, the overall corner point distance value and the fracture index can also be constructed through other basic mathematical operations, which will not be limited and elaborated here.

[0067] The greater the fracture index of the regional block, the greater the possibility that it is located at the fracture position of the long bone. The regional block corresponding to the fracture index greater than the preset fracture threshold is recorded as the fracture regional block. The Canny operator is used to perform edge detection on the area formed by the fracture regional blocks in the broken bone area to obtain the fracture edge of the broken bone area.

[0068] It should be noted that in the embodiment of the present invention, the Norm function is used for normalization processing. In the embodiment of the present invention, other normalization methods can also be selected, such as function transformation, maximum-minimum normalization and other normalization methods, which will not be limited here; the preset fracture threshold takes the empirical value of 0.8, and the implementer can set it according to the specific situation.

[0069] Step S3: According to the position and sharpness of the upper edge pixel points on the upper fracture edge, select the upper target base point on the upper fracture edge; according to the position of the upper target base point on the upper fracture edge and the relative size of the widths of the upper broken bone area and the lower broken bone area, select the lower target base point from the lower fracture edge.

[0070] The key to reducing the displacement of fractured bones is to set the reduction path of the fracture. A correct and short reduction path helps to minimize the damage to other human tissues during fracture reduction.

[0071] This solution is for reducing the displacement of fractured bones with complex fracture surfaces. A complex fracture surface means that after the fracture, the upper fracture opening has a certain degree of downward displacement, forming an upper and lower bone dislocation state. Usually, the lower fractured bone is fixed, and the upper fractured bone is moved to achieve fracture reduction. When setting the reduction path, it is necessary to avoid the situation of bone collision due to the complex fracture surface.

[0072] To avoid bone collision during fracture reduction, it is necessary to analyze the edge contour of the fracture opening. The upper fractured bone is in a convex-down state, and the lower fractured bone is in a convex-up state. If a pair of base points are arbitrarily located on the upper edge of the upper fracture opening and the lower edge of the lower fracture opening in the upper fracture area to set the reduction path, the convex-down part of the upper fractured bone will collide with the convex-up part of the lower fractured bone. This solution selects the upper target base point and the lower target base point based on the contour trends of the upper and lower fracture edges for setting the fracture reduction path.

[0073] The method for obtaining the upper target base point is as follows:

[0074] Considering that only the upper fractured bone is displaced during fracture reduction, when selecting the upper target base point, it is necessary to avoid collision between the upper fractured bone and the lower fractured bone during the movement of the upper fractured bone. The fracture opening of the upper fractured bone is wider at the top and narrower at the bottom. To avoid collision, it is only necessary to ensure that the lowest point of the narrower part of the upper fractured bone does not collide with the lower fractured bone. Therefore, select the edge pixel point on the upper fracture edge that is closest to the bottom of the image and the sharpest as the upper target base point.

[0075] In the embodiment of the present invention, the method for obtaining the upper target base point includes: arbitrarily selecting an edge pixel point on the upper fracture edge as an example point, on the upper fracture edge, denoting the adjacent previous edge pixel point and the adjacent next edge pixel point of the example point as the adjacent analysis points of the example point; denoting the line segment connecting the example point and each of its adjacent analysis points as the analysis edge of the example point; taking the angle between the two analysis edges of the example point as the sharpness index of the example point; obtaining the base point judgment index of the corresponding edge pixel point according to the ordinate of the pixel coordinates of each edge pixel point on the upper fracture edge and the sharpness index; the ordinate and the base point judgment index are in a positive correlation relationship, and the sharpness index and the base point judgment index are in a negative correlation relationship; denoting the edge pixel point corresponding to the largest base point judgment index on the upper fracture edge as the upper target base point.

[0076] Since the positive direction of the vertical axis of the image coordinate system is vertically downward, the larger the ordinate of the pixel coordinates of the upper-edge pixel points on the upper fracture edge, the greater the possibility that the edge pixel points are at the lowest point of the upper fracture area, and the greater the possibility that the edge pixel points are selected as the upper target base points. In this embodiment, the sharpness of the example points is measured by the angle between the two analysis edges of the example points; if the angle between the two analysis edges of the example points is smaller, it indicates that the sharpness of the position of the example points is more obvious, and the greater the possibility that the example points are selected as the upper target base points. Therefore, the ordinate of the pixel coordinates of the example points is positively correlated with the base point judgment index, and the sharpness index is negatively correlated with the base point judgment index.

[0077] In the embodiment of the present invention, the ratio of the ordinate of the pixel coordinates of the example points to the sharpness index is used as the base point judgment index of the example points. The greater the base point judgment index of the example points, the greater the possibility that the example points are selected as the upper target base points. The edge pixel point corresponding to the largest base point judgment index on the upper fracture edge is recorded as the upper target base point.

[0078] It should be noted that the edge pixel points at both end positions of the upper fracture edge do not participate in the selection of the upper target base points; the remaining edge pixel points on the upper fracture edge except the end positions are the intersection points of its two analysis edges, and the value range of the angle between the two analysis edges of the edge pixel points is .

[0079] The method for obtaining the lower target base points is as follows:

[0080] The cross-section of the upper broken bone basically coincides with the cross-section of the lower broken bone, but there will also be phenomena such as a small amount of broken bone falling off, resulting in poor accuracy of the lower target base points determined based on the contour consistency of the fracture edges of the upper and lower broken bones. Long bone fractures usually occur in the middle of the long bone, and the width at the middle position of the long bone is relatively close, and the widths of the two long bone fracture sites are approximately equal at a certain distance.

[0081] In the embodiment of the present invention, the method for obtaining the lower target base point includes: for each broken bone region, performing a morphological operation on the broken bone region to obtain the skeleton line of the broken bone region; respectively drawing perpendicular lines to each pixel point on the skeleton line, and connecting the two intersection points of the perpendicular line of each pixel point on the skeleton line with the edge of the broken bone region to obtain a line segment, which is denoted as the width line segment of each pixel point on the skeleton line; calculating the average value of the absolute value of the difference between the length of the width line segment of each pixel point on the skeleton line and the width line segment of its adjacent analysis point as the width difference value of each pixel point on the skeleton line; marking the pixel points on the skeleton line with a width difference value less than the preset difference threshold as the width judgment points of the broken bone region; marking the width line segment of the width judgment point corresponding to the maximum value in the ordinate of the pixel coordinates of the width judgment points in the upper broken bone region as the upper long bone width line segment; marking the width line segment of the width judgment point corresponding to the minimum value in the ordinate of the pixel coordinates of the width judgment points in the lower broken bone region as the lower long bone width line segment; obtaining the perpendicular point of the upper target base point on the upper long bone width line segment, which is denoted as the upper target point; selecting a lower target point from the lower long bone width line segment, and the distance between the lower target point and the left endpoint of the lower long bone width line segment is equal to the distance between the upper target point and the left endpoint of the upper long bone width line segment; drawing a perpendicular line to the lower target point on the lower long bone width line segment, and marking the intersection point of the perpendicular line of the lower target point and the edge of the lower fracture as the suspected lower target base point; for the edge pixel points within the preset local range of the lower target base point on the edge of the lower fracture, taking the edge pixel point corresponding to the maximum base point judgment index as the lower target base point.

[0082] If the width difference value of each pixel point on the skeleton line is smaller, it indicates that the bone width at the position of this pixel point of the broken bone is relatively close, and the possibility of using this pixel point to measure the bone width of the broken bone region is greater. Since the positive direction of the vertical axis of the image coordinate system is vertically downward, among the width judgment points in the upper broken bone region, the width judgment point with a larger ordinate is closer to the fracture position, and the accuracy for determining the lower target base point is higher.

[0083] Therefore, the width line segment of the width judgment point corresponding to the maximum value in the ordinate of the pixel coordinates of the width judgment points in the upper broken bone region is denoted as the upper long bone width line segment; the width line segment of the width judgment point corresponding to the minimum value in the ordinate of the pixel coordinates of the width judgment points in the lower broken bone region is denoted as the lower long bone width line segment.

[0084] Since the contours of the edges of the upper and lower broken bone fractures are consistent, the distances between the upper target point and the lower target point and the same-side edges of the broken bone region are relatively close. Therefore, the distance between the lower target point and the left endpoint of the lower long bone width line segment is equal to the distance between the upper target point and the left endpoint of the upper long bone width line segment. Among them, the left endpoint can be replaced with the right endpoint.

[0085] A perpendicular line to the line segment of the width of the lower long bone is drawn through the lower target point. The intersection point of the perpendicular line of the lower target point and the edge of the lower fracture is denoted as the suspected lower target base point, and the lower target base point is within the local range of the suspected lower target base point. The contours of the edges of the upper and lower fractured bones are consistent. The edge pixel point with the largest base point judgment index within the local range of the suspected lower target base point is taken as the lower target base point. Thus, the lower target base point is determined based on the width feature of the long bone.

[0086] It should be noted that in the embodiment of the present invention, a thinning algorithm is selected to obtain the skeleton line of the fractured bone area; the preset difference threshold takes an empirical value of 5, and the implementer can set it according to specific circumstances. Each edge pixel point on the lower fracture edge and its two left and two right edge pixel points form the preset local range of each pixel point on the lower fracture edge. For the two endpoints of the width line segment, the endpoint corresponding to the minimum value of the abscissa of the pixel coordinates is denoted as the left endpoint, and the endpoint corresponding to the maximum value of the abscissa of the pixel coordinates is denoted as the right endpoint. Among them, the thinning algorithm is a well-known technology to those skilled in the art and will not be elaborated here.

[0087] It should be noted that the method for obtaining the base point judgment index of the edge pixel points on the lower fracture edge is the same as that of the edge pixel points on the upper fracture edge; the method for obtaining the adjacent analysis points of each pixel point on the skeleton line is the same as that of the adjacent analysis points of the edge pixel points on the fracture edge, that is, the adjacent previous pixel point and the adjacent next pixel point of each pixel point on the skeleton line of the fractured bone area are denoted as the adjacent analysis points of each pixel point on the skeleton line.

[0088] Step S4: Determine the reduction path of the long bone fracture according to the relative positions of the upper target base point and the lower target base point, and the direction trend of the upper fracture edge; move the upper fractured bone area along the reduction path to assist in reducing the long bone fracture.

[0089] When setting the reduction path of the fracture, the cross-section situation of the long bone needs to be considered. For a more complex cross-section, in order to avoid collisions during the fracture reduction process, multiple paths need to be set for connection to generate a composite reduction path. Since the method of fixing the lower fractured bone and moving the upper fractured bone is adopted during fracture reduction, the path formed by moving the upper target base point to the position of the lower target base point is used as the fracture reduction path.

[0090] Fracture reduction needs to consider the edge contour of the fracture at the position of the upper target base point on the fractured bone to avoid collisions during the fracture reduction process. As Figure 2 shown, Figure 2The upper broken bone is slightly offset downward and to the lower right of the lower broken bone, and the abscissa of the pixel coordinates of the upper target base point is greater than the abscissa of the pixel coordinates of the lower target base point; during the fracture reduction process, the collision occurs at the lower left side of the upper broken bone and the upper right side of the lower broken bone. The setting of the first path of the reduction path needs to be based on the edge distribution of the lower left side of the upper broken bone. If the upper broken bone is offset downward and to the left, the collision during the reduction process occurs at the lower right side of the upper broken bone and the upper left side of the lower broken bone. Therefore, according to the relative position of the upper target base point and the lower target base point, and the direction trend of the upper fracture edge, the reduction path is determined, and thus the reduction midpoint is obtained.

[0091] In the embodiment of the present invention, the method for obtaining the reduction midpoint is as follows: if the abscissa of the pixel coordinates of the upper target base point is greater than the abscissa of the pixel coordinates of the lower target base point, the edge between the left endpoint of the upper fracture edge and the upper target base point is recorded as the direction analysis edge, and the edge between the right endpoint of the lower fracture edge and the lower target base point is recorded as the collision analysis edge; if the abscissa of the pixel coordinates of the upper target base point is less than the abscissa of the pixel coordinates of the target base point, the edge between the right endpoint of the upper fracture edge and the upper target base point is recorded as the direction analysis edge, and the edge between the left endpoint of the lower fracture edge and the lower target base point is recorded as the collision analysis edge; the edge pixels on the direction analysis edge are linearly fitted to obtain a direction line; a length analysis line segment of the lower target base point is obtained, and there is an intersection point between the length analysis line segment and the collision analysis edge; the intersection point of the direction line and the length analysis line segment is recorded as the reduction midpoint.

[0092] The direction analysis edge is a partial edge of the upper fracture edge that needs to be referred to during the fracture reduction process and is used to determine the direction of the first path of the reduction path; the collision analysis edge is the edge where a collision may occur during the fracture reduction process and is used to determine the length of the first path of the reduction path. When there is only one edge pixel between the length analysis line segment and the collision analysis edge, the upper broken bone area moves linearly from the reduction midpoint to the lower target base point and just does not collide with the lower fracture edge. Therefore, the straight line from the upper target base point to the reduction midpoint is used as the first path.

[0093] It should be noted that since long bone fractures will cause the upper broken bone to shift relative to the lower broken bone, the abscissas of the pixel coordinates of the upper target base point and the lower target base point will not be equal; in this embodiment, the least squares method is used to linearly fit the edge pixels on the direction analysis edge. Among them, the least squares method is a well-known technology to those skilled in the art and will not be elaborated here.

[0094] When setting the second path, there is no need to consider the problem of collision between the upper and lower broken bones. In order to reduce the damage to other human tissues during the bone fracture reduction process, it is only necessary to translate the upper broken bone based on the midpoint of the path and the lower target base point. Since the straight line between two points is the shortest, the straight line from the reduction midpoint to the lower target base point is used as the second path.

[0095] The reduction path is composed of the first path and the second path. The midpoint of the path is the end point of the first path and also the starting point of the second path. The upper target base point is the starting point of the reduction path, and the lower target base point is the end point of the reduction path.

[0096] The specific process of long bone fracture reduction is as follows: The lower broken bone area remains stationary, and the upper target base point in the upper broken bone area is moved along the reduction path towards the lower target base point. While the upper target base point is moving, the upper broken bone area moves together with the upper target base point, that is, the moving paths of all pixel points in the upper broken bone area are the same length as the reduction path, and the directions of each segment of the moving path and the reduction path are parallel.

[0097] This solution considers the position distribution of the two parts of the broken bone area and the distribution of the fracture edge contour to set the reduction path of long bone fracture. On the premise of ensuring no collision during the bone fracture reduction process, the length of the reduction path is minimized to reduce the damage to other human tissues during the bone fracture reduction process and improve the reduction effect of long bone fracture.

[0098] So far, the present invention is completed.

[0099] Embodiment 2:

[0100] The present invention provides a long bone fracture reduction system based on image processing. Please refer to Figure 3 , which shows the system structure diagram of a long bone fracture reduction system based on image processing provided by an embodiment of the present invention. The system includes:

[0101] A data acquisition module 510, configured to acquire long bone fracture images of the fracture site of a patient;

[0102] A fracture edge acquisition module 520, configured to perform region segmentation on the long bone fracture image to obtain an upper broken bone area and a lower broken bone area; acquire the upper fracture edge and the lower fracture edge;

[0103] A target base point screening module 530, configured to select an upper target base point on the upper fracture edge according to the position and sharpness of the upper edge pixel points of the upper fracture edge; select a lower target base point from the lower fracture edge according to the position of the upper target base point on the upper fracture edge and the relative size of the widths of the upper broken bone area and the lower broken bone area;

[0104] The auxiliary reset module 540 is configured to determine the reset path of the long bone fracture according to the relative position between the upper target base point and the lower target base point, and the direction trend of the upper fracture edge; move the upper fractured bone area along the reset path to assist in the reset of the long bone fracture.

[0105] It should be noted that for the device provided in the above embodiment, only the division of the above functional modules is used for illustration. In practical applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the computer device is divided into different functional modules to complete all or part of the functions described above. In addition, the long bone fracture reduction system based on image processing and the long bone fracture reduction method embodiment provided in the above embodiment belong to the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0106] Embodiment 3:

[0107] Figure 4 This is a schematic diagram of a computer device for a long bone fracture reduction device based on image processing provided by an embodiment of the present invention. Exemplarily, as Figure 4 shown, the computer device includes: a memory 601, a processor 602, and a computer program 603 stored in the memory 601 and running on the processor 602. When the processor 602 executes the computer program 603, the computer device can execute any of the long bone fracture reduction methods based on image processing introduced above.

[0108] In addition, an embodiment of the present application also protects a device, which may include a memory and a processor. Among them, the memory stores executable program code, and the processor is used to call and execute the executable program code to execute a long bone fracture reduction method provided by an embodiment of the present application.

[0109] In this embodiment, the device can be divided into functional modules according to the above method examples. For example, it can correspond to each functional module, or two or more functions can be integrated into one processing module. The above integrated module can be implemented in hardware form. It should be noted that the division of modules in this embodiment is illustrative, only a logical function division, and there may be other division methods in actual implementation.

[0110] It should be understood that the device provided in this embodiment is used to execute the above long bone fracture reduction method based on image processing, so the same effect as the above implementation method can be achieved.

[0111] In the case of adopting an integrated unit, the device may include a processing module and a storage module. Among them, when the device is applied to a device, the processing module can be used to control and manage the actions of the device. The storage module can be used to support the device to execute mutual program codes, etc.

[0112] Among them, the processing module can be a processor or a controller, which can implement or execute various exemplary logic blocks, modules and circuits included in combination with the disclosure of the present application. The processor can also be a combination that realizes computing functions, such as including a combination of one or more microprocessors, a combination of digital signal processing (DSP) and a microprocessor, etc. The storage module can be a memory.

[0113] It should be noted that: the above sequence of embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0114] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. The key point of each embodiment is to illustrate the differences from other embodiments.

[0115] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A long bone fracture reduction system based on image processing, characterized in that: The system includes: a data acquisition module for acquiring a long bone fracture image of a patient's fracture site; A fracture edge acquisition module is used to segment the long bone fracture image into an upper fracture area and a lower fracture area; and to acquire an upper fracture edge and a lower fracture edge; The target base point screening module is used to select the upper target base point on the upper fracture edge according to the position and sharpness of the upper edge pixel point of the upper fracture edge; select the lower target base point from the lower fracture edge according to the position of the upper target base point on the upper fracture edge and the relative size of the width of the upper fracture area and the lower fracture area; The auxiliary reduction module is used to obtain the reduction midpoint according to the relative position of the upper target base point and the lower target base point, and the direction of the upper fracture edge; the straight line from the upper target base point to the reduction midpoint is used as the first path; the straight line from the reduction midpoint to the lower target base point is used as the second path; the reduction path of the long bone fracture is determined by the first path and the second path; the upper fractured bone area is moved along the reduction path to assist in the reduction of the long bone fracture; The obtaining of the reset midpoint comprises: If the abscissa of the pixel coordinates of the upper target base point is greater than the abscissa of the pixel coordinates of the lower target base point, the edge between the left end point of the upper fracture edge and the upper target base point is recorded as the direction analysis edge, and the edge between the right end point of the lower fracture edge and the lower target base point is recorded as the collision analysis edge; if the abscissa of the pixel coordinates of the upper target base point is less than the abscissa of the pixel coordinates of the lower target base point, the edge between the right end point of the upper fracture edge and the upper target base point is recorded as the direction analysis edge, and the edge between the left end point of the lower fracture edge and the lower target base point is recorded as the collision analysis edge; Perform straight line fitting on the edge pixel points on the direction analysis edge to obtain the direction line; obtain the length analysis line segment of the lower target base point, where there is an intersection between the length analysis line segment and the collision analysis edge; and record the intersection of the direction line and the length analysis line segment as the reset midpoint.

2. According to the long bone fracture reduction system based on image processing in claim 1, the obtaining of the upper fracture edge and the lower fracture edge comprises: Obtaining corner points in long bone fracture images; The upper fractured bone region and the lower fractured bone region are recorded as fractured bone regions, and each fractured bone region is divided into different regional blocks; For each area block, any two corner points in the area block form a corner point pair, and the average of the distances between the two corner points in all the corner point pairs of the area block is used as the overall value of the corner point distance of the area block; according to the total number of corner points in the area block and the overall value of the corner point distance, the fracture index of the area block is obtained; the total number of corner points in the area block is positively correlated with the fracture index, and the overall value of the corner point distance is negatively correlated with the fracture index; For each area block of the broken bone region, the area block corresponding to the fracture index greater than the preset fracture threshold is recorded as the fracture area block; the edge formed by the area of ​​all the fracture area blocks in the broken bone region is detected and the edge obtained is taken as the fracture edge; The fracture edge of the upper fractured bone region was recorded as the upper fracture edge, and the fracture edge of the lower fractured bone region was recorded as the lower fracture edge.

3. The long bone fracture reduction system based on image processing according to claim 1, characterized in that: The step of selecting an upper target base point on the upper fracture edge includes: An edge pixel point is randomly selected from the upper fracture edge and recorded as an example point. On the upper fracture edge, the previous edge pixel point and the next edge pixel point adjacent to the example point are recorded as adjacent analysis points of the example point. The line segment obtained by connecting the example point and each of its adjacent analysis points is recorded as the analysis edge of the example point; the angle between the two analysis edges of the example point is taken as the sharpness index of the example point; According to the ordinate of the pixel coordinates of each edge pixel point on the upper fracture edge and the sharpness index, a base point judgment index corresponding to the edge pixel point is obtained; the ordinate and the base point judgment index are positively correlated, and the sharpness index and the base point judgment index are negatively correlated; The edge pixel point corresponding to the largest base point judgment index on the upper fracture edge is recorded as the upper target base point.

4. The long bone fracture reduction system based on image processing according to claim 3, characterized in that: The step of selecting a lower target base point from the lower fracture edge includes: For each broken bone region, a morphological operation is performed on the broken bone region to obtain a skeleton line of the broken bone region; a perpendicular line is drawn to each pixel point on the skeleton line, and a line segment is obtained by connecting the perpendicular line of each pixel point on the skeleton line with two intersection points of the edge of the broken bone region, and the line segment is recorded as the width line segment of each pixel point on the skeleton line; Calculate the average of the absolute values ​​of the differences between the lengths of the width segments of each pixel point on the skeleton line and its adjacent analysis point as the width difference value of each pixel point on the skeleton line; record the pixel points on the skeleton line whose width difference values ​​are less than a preset difference threshold as width judgment points of the broken bone area; The width line segment of the width judgment point corresponding to the maximum value in the ordinate of the pixel coordinates of the width judgment point in the upper broken bone region is recorded as the upper long bone width line segment; the width line segment of the width judgment point corresponding to the minimum value in the ordinate of the pixel coordinates of the width judgment point in the lower broken bone region is recorded as the lower long bone width line segment; Obtain a vertical point of the upper target base point on the upper long bone width line segment, recorded as the upper target point; select a lower target point from the lower long bone width line segment, the distance between the lower target point and the left end point of the lower long bone width line segment is equal to the distance between the upper target point and the left end point of the upper long bone width line segment; Draw perpendicular lines to the upper and lower target points of the lower long bone width segment, and record the intersection of the perpendicular line to the lower target point and the edge of the lower fracture as the suspected lower target base point; for edge pixel points within a preset local range of the upper and lower target base points of the lower fracture edge, take the edge pixel point corresponding to the largest base point judgment index as the lower target base point.

5. The long bone fracture reduction system based on image processing according to claim 1, characterized in that: The method of performing regional segmentation on the long bone fracture image to obtain an upper fractured bone region and a lower fractured bone region includes: The maximum inter-class variance method is used to obtain the segmentation threshold of the long bone fracture image; in the long bone fracture image, the connected domain formed by the pixel points whose gray value is greater than the segmentation threshold is recorded as the broken bone area; there are two broken bone areas; Obtain the geometric center of the broken bone area; for the vertical coordinates of the pixel coordinates of the geometric centers of the two broken bone areas, the broken bone area corresponding to the smallest vertical coordinate is recorded as the upper broken bone area, and the broken bone area corresponding to the largest vertical coordinate is recorded as the lower broken bone area.

6. The long bone fracture reduction system based on image processing according to claim 1, characterized in that: The moving the upper broken bone region along the reduction path comprises: moving an upper target base point on the upper broken bone region along the reduction path.

7. The long bone fracture reduction system based on image processing according to claim 2, characterized in that: The algorithm for obtaining corner points in long bone fracture images is the Harris corner point detection algorithm.

Images