Auxiliary detection method and system for lumbar osteoporosis based on X-ray films

Abstract

The present invention relates to the field of image processing technology, and in particular to an auxiliary detection method and system for lumbar osteoporosis based on X-ray films. The method comprises: obtaining a lumbar X-ray film image; determining a vertebral region in combination with the shape distribution of edge lines in the lumbar X-ray film image; obtaining a value of the degree of influence of each pixel point on the overlap of bones based on the difference in the position distribution of each pixel point and its similar pixel points; obtaining a difference index corresponding to each pixel point in the vertebral region according to the difference between the standard grayscale of each pixel point in the vertebral region and the normal vertebral region; obtaining a lumbar osteoporosis evaluation value by combining the influence degree value and the difference index, and then obtaining an accurate auxiliary detection result of lumbar osteoporosis. The present invention improves the accuracy of the judgment result of suspected osteoporosis.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to an auxiliary detection method and system for lumbar osteoporosis based on X-ray films. Background Art

[0002] Osteoporosis is a disease characterized by decreased bone density and destruction of bone microstructure. Osteoporosis makes bones more susceptible to injury and fractures take longer to heal than normal bones, affecting daily life and activities. X-rays are a common tool for diagnosing osteoporosis and can help to initially assess the manifestations of osteoporosis, such as vertebral compression fractures or sparse trabeculae.

[0003] However, when analyzing the manifestations of lumbar osteoporosis through X-rays, the lumbar vertebrae often overlap with other bone tissues, affecting the performance of bone density, making it difficult to directly obtain accurate auxiliary test results of lumbar osteoporosis in the person to be tested through the grayscale uniformity of the lumbar part in the X-ray, affecting the subsequent judgment results. Summary of the invention

[0004] In order to solve the problem that the auxiliary detection results of lumbar osteoporosis of the person to be detected obtained by the existing method have low accuracy, the purpose of the present invention is to provide a method and system for auxiliary detection of lumbar osteoporosis based on X-ray films, and the technical scheme adopted is as follows:

[0005] In a first aspect, the present invention provides an auxiliary detection method for lumbar osteoporosis based on X-ray films, the method comprising the following steps:

[0006] Obtaining a lumbar spine X-ray image of the person to be tested;

[0007] The vertebral area is determined by combining the shape distribution of the edge line in the lumbar X-ray image; the similar pixel points corresponding to each pixel point in the vertebral area are screened according to the difference between the local grayscale distribution of each pixel point and other pixel points in the vertebral area; the influence degree value of each pixel point affected by the bone overlap is obtained based on the difference in the position distribution of each pixel point and its similar pixel points; the difference index corresponding to each pixel point in the vertebral area is obtained according to the difference between the standard grayscale of each pixel point in the vertebral area and the normal vertebral area;

[0008] The impact degree value and the difference index are combined to obtain a lumbar osteoporosis evaluation value; based on the lumbar osteoporosis evaluation value, an auxiliary detection result of lumbar osteoporosis of the person to be detected is obtained.

[0009] Preferably, the step of determining the vertebral area by combining the shape distribution of the edge lines in the lumbar X-ray image includes:

[0010] Divide each edge line in the lumbar spine X-ray image into several sub-segments;

[0011] Determine the possible value of each sub-line segment being an arc shape according to the slope difference between adjacent pixel points on each sub-line segment, wherein the slope difference is positively correlated with the possible value of the arc shape; determine the sub-line segment corresponding to the possible value of the arc shape being greater than a preset first possible threshold as a suspected vertebral foramen edge line;

[0012] Determine the possible value of each suspected vertebral foramen edge line as a vertebral foramen edge line according to the overall distribution of the slope of the pixel points on each suspected vertebral foramen edge line; determine the suspected vertebral foramen edge line corresponding to the possible value of the vertebral foramen edge line greater than a preset second possible threshold as the vertebral foramen edge line;

[0013] The vertebral body area was obtained by connecting the edge lines of all vertebral foramina.

[0014] Preferably, the step of screening similar pixels corresponding to each pixel in the vertebral body region according to the difference between the local grayscale distribution of each pixel in the vertebral body region and other pixels includes:

[0015] According to the difference in grayscale values ​​between the candidate pixel and the pixel points at corresponding positions in the neighborhood of each other pixel point in the vertebral region, a similarity index between the candidate pixel point and each other pixel point in the vertebral region is obtained; the candidate pixel point is any pixel point in the vertebral region;

[0016] Based on the similarity index between each pixel point in the vertebral body region and each other pixel point in the vertebral body region, similar pixel points corresponding to each pixel point in the vertebral body region are screened.

[0017] Preferably, the screening of similar pixel points corresponding to each pixel point in the vertebral region based on a similarity index between each pixel point in the vertebral region and each other pixel point in the vertebral region includes:

[0018] For a candidate pixel point, a pixel point whose similarity index with the candidate pixel point is greater than a preset similarity threshold is taken as a similar pixel point corresponding to the candidate pixel point.

[0019] Preferably, obtaining the influence degree value of each pixel point due to the overlap of bones based on the difference in position distribution between each pixel point and similar pixels thereof comprises:

[0020] According to the position coordinates of the candidate pixel point and its corresponding similar pixel point in the lumbar spine X-ray image, the fitting values ​​corresponding to the candidate pixel point and its corresponding similar pixel point are obtained;

[0021] According to the difference between the horizontal coordinates of the candidate pixel point and its corresponding similar pixel point and the corresponding fitting value, the degree of influence of the candidate pixel point due to bone overlap is obtained, and the difference between the horizontal coordinate and the corresponding fitting value is negatively correlated with the degree of influence.

[0022] Preferably, obtaining the fitting values ​​corresponding to the candidate pixel points and their corresponding similar pixel points according to their position coordinates in the lumbar spine X-ray image includes:

[0023] The candidate pixel and all its corresponding similar pixels are recorded as the first pixel;

[0024] Obtaining position coordinates of all first pixel points in the lumbar spine X-ray image, and exchanging the abscissa and ordinate in the position coordinates of each first pixel point to obtain corresponding new position coordinates;

[0025] Based on the new position coordinates of all the first pixel points, curve fitting is performed on all the first pixel points to obtain a fitting curve corresponding to the candidate pixel point;

[0026] The ordinate on the fitting curve is used as the fitting value corresponding to the first pixel point.

[0027] Preferably, obtaining the difference index corresponding to each pixel point in the vertebral body region according to the difference between each pixel point in the vertebral body region and the standard grayscale of the normal vertebral body region includes:

[0028] The difference between the gray value of the candidate pixel and the standard gray value of the normal vertebral area is used as the difference index corresponding to the candidate pixel.

[0029] Preferably, the combining the impact degree value and the difference index to obtain the lumbar osteoporosis evaluation value includes:

[0030] Obtaining the osteoporosis degree value of each vertebral area according to the influence degree value of the bone overlap of each pixel point in each vertebral area and the corresponding difference index, wherein the influence degree value is negatively correlated with the osteoporosis degree value, and the difference index is positively correlated with the osteoporosis degree value;

[0031] The maximum value of the osteoporosis degree values ​​of all vertebral areas is determined as the lumbar osteoporosis evaluation value.

[0032] Preferably, the auxiliary test result of lumbar osteoporosis of the person to be tested is obtained based on the lumbar osteoporosis evaluation value, including:

[0033] When the lumbar vertebra osteoporosis evaluation value is greater than the preset osteoporosis evaluation threshold, it is determined that the lumbar vertebra of the person to be tested has suspected osteoporosis;

[0034] When the lumbar osteoporosis evaluation value is less than or equal to a preset osteoporosis evaluation threshold, it is determined that the lumbar spine of the person to be tested does not have suspected osteoporosis.

[0035] In a second aspect, the present invention provides an X-ray-based lumbar osteoporosis auxiliary detection system, the system comprising:

[0036] An image acquisition module is used to obtain a lumbar spine X-ray image of the person to be tested;

[0037] A calculation module is used to determine the vertebral area in combination with the shape distribution of the edge line in the lumbar X-ray image; screen similar pixels corresponding to each pixel in the vertebral area according to the difference between the local grayscale distribution of each pixel and other pixels in the vertebral area; obtain the value of the degree of influence of the bone overlap on each pixel based on the difference in the position distribution of each pixel and its similar pixels; obtain the difference index corresponding to each pixel in the vertebral area according to the difference between the standard grayscale of each pixel in the vertebral area and the normal vertebral area;

[0038] The judgment module is used to obtain a lumbar osteoporosis evaluation value by combining the impact degree value and the difference index; and obtain an auxiliary detection result of lumbar osteoporosis of the person to be detected based on the lumbar osteoporosis evaluation value.

[0039] The present invention has at least the following beneficial effects:

[0040] The present invention firstly extracts the vertebral area in the lumbar spine X-ray image of the person to be tested in combination with the shape characteristics of the vertebral area, and by analyzing the similarity of the grayscale distribution of different pixel points, it can effectively avoid the problem that the grayscale distribution of local positions with similar structures caused by lumbar curvature is inconsistent, thereby affecting the accuracy of the analysis result. Under normal circumstances, each vertebral structure on the lumbar spine should have similar characteristics. The present invention uses the regularity of the position distribution between pixel points with similar grayscale distribution to judge whether the similar performance of these similar pixel points is caused by the similarity of the lumbar vertebral structure, and obtains the difference index corresponding to each pixel point in the vertebral area, and then evaluates the degree of influence of bone overlap on each pixel point, and obtains the value of the degree of influence of bone overlap on each pixel point, and further combines the value of the degree of influence of bone overlap on the pixel point and the corresponding difference index to obtain the auxiliary detection result of lumbar osteoporosis of the person to be tested, and the subsequent doctor makes another judgment based on the result, thereby improving the accuracy of the judgment result and effectively ensuring the detection result of lumbar osteoporosis of the person to be tested. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0042] Figure 1 A flowchart of an X-ray-based auxiliary detection method for lumbar osteoporosis provided by an embodiment of the present invention;

[0043] Figure 2 A flowchart of a method for obtaining an influence degree value and a difference index provided by an embodiment of the present invention;

[0044] Figure 3 A schematic diagram of a vertebral region provided by an embodiment of the present invention;

[0045] Figure 4 This is a structural block diagram of an X-ray-based lumbar osteoporosis auxiliary detection system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0046] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the X-ray-based lumbar osteoporosis auxiliary detection method and system proposed by the present invention are described in detail as follows in combination with the accompanying drawings and preferred embodiments.

[0047] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0048] The specific scheme of the X-ray-based lumbar osteoporosis auxiliary detection method and system provided by the present invention is described in detail below with reference to the accompanying drawings.

[0049] Example of auxiliary detection method for lumbar osteoporosis based on X-ray film:

[0050] The specific scenario targeted by this embodiment is: when using lumbar X-rays to analyze the degree of lumbar osteoporosis, the lumbar bones often overlap with other bones, resulting in a large difference between the lumbar bone density shown in the X-ray and the actual bone density, making the evaluation result of lumbar osteoporosis inaccurate. This embodiment uses the structural and shape similarity characteristics between vertebral bodies in the lumbar spine, analyzes the distribution relationship between lumbar positions with similar grayscale expressions, identifies the lumbar positions affected by the overlap of other bones, and effectively reduces the impact of bone overlap on the evaluation results of lumbar osteoporosis.

[0051] This embodiment proposes an auxiliary detection method for lumbar osteoporosis based on X-ray films, such as Figure 1 As shown, the X-ray-based auxiliary detection method for lumbar osteoporosis in this embodiment includes the following steps:

[0052] Step S1, obtaining a lumbar spine X-ray image of the person to be tested.

[0053] First, an X-ray image of the lumbar spine area of ​​the person to be tested is taken, and then the X-ray image of the lumbar spine area is input into the trained semantic segmentation network. The output of the semantic segmentation network is the X-ray image of the lumbar spine of the person to be tested, which means that the influence of other areas on subsequent detection results is excluded. The semantic segmentation network in this embodiment is used to extract the X-ray image of the lumbar spine area in the collected image, excluding the influence of other areas. The semantic segmentation network and the training process of the semantic segmentation network are both existing technologies and will not be described in detail here.

[0054] At this point, the lumbar spine X-ray images of the person to be tested have been collected.

[0055] Step S2, combining the shape distribution of the edge line in the lumbar X-ray image, determining the vertebral area; screening similar pixels corresponding to each pixel in the vertebral area according to the difference between the local grayscale distribution of each pixel and other pixels in the vertebral area; obtaining the value of the degree of influence of bone overlap on each pixel based on the difference in the position distribution of each pixel and its similar pixels; obtaining the difference index corresponding to each pixel in the vertebral area according to the difference between the standard grayscale of each pixel in the vertebral area and the normal vertebral area.

[0056] The vertebrae of the lumbar spine are similar in shape and structure, that is, the same relative position on different vertebrae in the X-ray should have similar grayscale distribution. When other bone tissues overlap with the lumbar spine, the density of other bones will affect the grayscale performance of the overlapping part of the lumbar spine, destroying this similarity. Therefore, based on the relative position relationship between the lumbar vertebrae with similar grayscale performance, it is possible to analyze whether there is other bone overlap at each lumbar position, thereby reducing the impact of the overlapping part on the osteoporosis assessment results.

[0057] Step S21, determining the vertebral body region in combination with the shape distribution of the edge lines in the lumbar spine X-ray image.

[0058] Lumbar osteoporosis is mainly manifested in the vertebral body of the lumbar spine, so only the vertebral body needs to be analyzed. From the lumbar X-ray image, it can be seen that the edge of the vertebral foramen between each vertebra is arc-shaped, while the rest of the vertebral body is relatively straight, so the edge of the vertebral foramen can be identified through this feature and each vertebral area can be extracted. Figure 3 As shown, this figure is a schematic diagram of a portion of the vertebral area.

[0059] First, the Canny edge detection algorithm is used to perform edge detection on the lumbar spine X-ray image to extract the edge lines in the lumbar spine X-ray image. Then, each edge line in the lumbar spine X-ray image is divided into a number of sub-segments; in this embodiment, the length of each sub-segment is equal, which is 20 pixels. In specific applications, the implementer can set it according to the specific situation.

[0060] Next, according to the slope difference between adjacent pixels on each sub-segment, the possible value of each sub-segment being an arc shape is determined, and the slope difference is positively correlated with the possible value of the arc shape; the sub-segment corresponding to the possible value of the arc shape when it is greater than the preset first possible threshold is determined as the suspected vertebral foramen edge line. According to the overall distribution of the slopes of the pixels on each suspected vertebral foramen edge line, the possible value of each suspected vertebral foramen edge line as a vertebral foramen edge line is determined; the suspected vertebral foramen edge line corresponding to the possible value of the vertebral foramen edge line when it is greater than the preset second possible threshold is determined as the vertebral foramen edge line. In this embodiment, the preset first possible threshold is 0.5, and the preset second threshold is 0.7. In specific applications, the implementer can set it according to specific circumstances.

[0061] In this embodiment, specific calculation formulas are given for the possible value of the sub-line segment being an arc shape and the possible value of the suspected vertebral foramen edge line being a vertebral foramen edge line. The possible value of the i-th sub-line segment being an arc shape and the possible value of the j-th suspected vertebral foramen edge line being a vertebral foramen edge line can be expressed as:

[0062] ;

[0063] ;

[0064] in, represents the possible value of the i-th sub-segment being an arc, represents the possible value of the jth suspected vertebral foramen edge line as the vertebral foramen edge line, U represents the number of pixels on the i-th sub-segment, represents the first The slope of the pixel, represents the slope of the u-th pixel on the i-th sub-segment, V represents the number of pixels on the j-th suspected vertebral foramen edge line, represents the slope of the vth pixel point on the jth suspected vertebral foramen edge line, represents the normalization function, Indicates the absolute value sign.

[0065] represents the slope difference between the u+1th pixel and the uth pixel on the i-th sub-segment, The cumulative sum of the slope differences between adjacent pixels on the ith sub-segment represents the cumulative sum. The larger the cumulative sum, the more the ith sub-segment is arc-shaped, that is, the greater the probability that the ith sub-segment is arc-shaped. It represents the sum of the slopes of all pixels on the jth suspected vertebral foramen edge line. The smaller the sum is, the closer the overall shape of the jth suspected vertebral foramen edge line is to a circle, and the more likely it is to be a vertebral foramen edge line, that is, the greater the possibility that the i-th sub-segment is in the shape of an arc.

[0066] By adopting the above method, multiple vertebral foramen edge lines can be screened out, and the vertebral foramen edge lines with adjacent endpoints are connected, and each obtained closed connected domain is used as a vertebral body region, that is, multiple vertebral body regions are obtained. This embodiment will be described below using any vertebral body region as an example, and the method provided in this embodiment can be used to process other vertebral body regions.

[0067] Step S22, based on the difference between the local grayscale distribution of each pixel point and other pixels in the vertebral region, similar pixels corresponding to each pixel point in the vertebral region are screened.

[0068] The structural characteristics of the pixel points at each position on the lumbar spine are often reflected in the grayscale distribution of the surrounding area, so it is necessary to analyze the grayscale distribution of the pixel points in the local area of ​​each lumbar pixel point. Lumbar osteoporosis is usually manifested as reduced bone density, that is, the grayscale value is low in the grayscale image of the X-ray film, and the bones at the edge of the lumbar vertebra are usually thicker, so the grayscale value is relatively large in the grayscale image of the X-ray film. Therefore, the grayscale distribution of the pixels in the neighborhood of each pixel point can reflect the structural characteristics of the location of the pixel point.

[0069] Next, this embodiment is described by taking any pixel point in the vertebral region as an example, and other pixel points can be processed by the method provided in this embodiment.

[0070] Specifically, any pixel point in the cone area is recorded as a candidate pixel point, and the similarity index between the candidate pixel point and each other pixel point in the cone area is obtained according to the difference in grayscale values ​​between the candidate pixel point and the pixel points at corresponding positions in the neighborhood of each other pixel point in the cone area.

[0071] In this embodiment, a specific calculation formula for the similarity index is given. The similarity index between the c-th pixel point and the d-th pixel point in the vertebral region can be expressed as:

[0072] ;

[0073] in, represents the similarity index between the cth pixel and the dth pixel in the vertebral region, represents the gray value of the rth pixel in the neighborhood of the cth pixel in the vertebral area, represents the gray value of the rth pixel in the neighborhood of the dth pixel in the vertebral area, R represents the number of pixels in the neighborhood of each pixel, To preset the first adjustment parameter, represents the normalization function, Indicates the absolute value sign.

[0074] The purpose of introducing the preset first adjustment parameter into the calculation formula of the similarity index is to prevent the denominator from being 0. In this embodiment, the preset first adjustment parameter is 0.01. In specific applications, the implementer can set it according to the specific situation. It represents the grayscale difference between the rth pixel in the neighborhood of the cth pixel and the rth pixel in the neighborhood of the dth pixel in the vertebral region. Characterizes the overall grayscale difference between all pixels in the neighborhood of the c-th pixel and all pixels in the neighborhood of the d-th pixel. The larger the value, the greater the local grayscale difference between the c-th pixel and the d-th pixel, the more dissimilar the distribution characteristics of the c-th pixel and the d-th pixel, that is, the smaller the similarity index between the c-th pixel and the d-th pixel in the vertebral region.

[0075] In this embodiment, for any pixel point: the neighborhood of the pixel point is to build a window of a preset size with the pixel point as the window center, and the other areas within the window area except the window center are regarded as the neighborhood of the pixel point. The preset size in this embodiment is ,In specific applications, implementers can set it according to specific circumstances.

[0076] It should be noted that for some pixels located at or close to the edge, some pixels may be missing in their neighborhood, and the grayscale values ​​of the missing pixels in their neighborhood are complemented with the average grayscale values ​​of the pixels in their neighborhood.

[0077] By using the above method, the similarity index between the candidate pixel and each other pixel in the vertebral region can be obtained, and the pixel whose similarity index with the candidate pixel is greater than the preset similarity threshold is used as the similar pixel corresponding to the candidate pixel. In this embodiment, the preset similarity threshold is 0.6, and in specific applications, the implementer can set it according to specific circumstances.

[0078] By adopting the above method, similar pixel points corresponding to the candidate pixel points can be screened out.

[0079] Step S23, based on the difference in position distribution between each pixel and its similar pixels, obtain the influence degree value of each pixel due to the overlap of bones.

[0080] Under normal circumstances, the same relative position of different vertebrae should have similar structural features, that is, the position relationship between pixels with similar structures presents a linear distribution. However, if bones overlap, this regularity will be destroyed, so that the positions with similar grayscale distribution are not truly structurally similar, that is, the relationship between these positions is irregular.

[0081] Based on the above characteristics, the candidate pixel point and all the similar pixel points corresponding to it are recorded as the first pixel point; the pixel point at the vertex of the lower left corner of the collected lumbar X-ray image is used as the coordinate origin, the horizontal right direction is used as the positive direction of the horizontal axis, and the vertical upward direction is used as the positive direction of the vertical axis to construct a rectangular coordinate system to obtain the position coordinates of all the first pixel points in the lumbar X-ray image, and the position coordinates include the horizontal coordinate and the vertical coordinate; it should be noted that: the shape of the lumbar X-ray image collected in this embodiment is a rectangle. The horizontal coordinate and the vertical coordinate in the position coordinate of each first pixel point are interchanged to obtain the corresponding new position coordinates; based on the new position coordinates of all the first pixel points, the least squares method is used to perform curve fitting on all the first pixel points, and the fitting curve obtained at this time is recorded as the fitting curve corresponding to the candidate pixel point; if the position distribution of these pixel points shows regularity, then the pixel point position should be highly consistent with the fitting curve. The vertical coordinate on the fitting curve is used as the fitting value corresponding to the corresponding first pixel point.

[0082] According to the difference between the horizontal coordinates of the candidate pixel point and its corresponding similar pixel point and the corresponding fitting value, the degree of influence of the candidate pixel point due to bone overlap is obtained, and the difference between the horizontal coordinate and the corresponding fitting value is negatively correlated with the degree of influence.

[0083] Among them, the negative correlation relationship means that the dependent variable will decrease as the independent variable increases, and the dependent variable will increase as the independent variable decreases. It can be a subtraction relationship, a division relationship, etc., which is determined by practical applications.

[0084] In this embodiment, a specific calculation formula for the influence degree of the bone overlap on the pixel point is given. The influence degree of the bone overlap on the c-th pixel point in the vertebral region can be expressed as:

[0085] ;

[0086] in, Indicates the degree of influence of bone overlap on the c-th pixel in the vertebral area. represents the total number of similar pixels corresponding to the c-th pixel in the vertebral area. represents the horizontal coordinate of the mth pixel point in the vertebral region and the similar pixel points corresponding to the cth pixel point, It represents the fitting value corresponding to the mth pixel point in the vertebral region and the similar pixels corresponding to the cth pixel point. To preset the second adjustment parameter, represents the normalization function, Indicates the absolute value sign.

[0087] The purpose of introducing the preset second adjustment parameter into the calculation formula of the impact degree value is to prevent the denominator from being 0. In this embodiment, the preset second adjustment parameter is 0.01. In specific applications, the implementer can set it according to the specific situation. Represents the overall difference between the c-th pixel point and the abscissa of the similar pixel points corresponding to the c-th pixel point and the corresponding fitting value. The larger the value is, the greater the overall difference is, that is, the smaller the degree of influence of bone overlap on the c-th pixel point in the vertebral area is.

[0088] By using the above method, the influence degree of each pixel point affected by the bone overlap can be obtained.

[0089] Step S24, obtaining a difference index corresponding to each pixel in the vertebral region according to the difference between each pixel in the vertebral region and the standard grayscale of a normal vertebral region.

[0090] Lumbar osteoporosis is a decrease in bone density, which is manifested as a decrease in the grayscale value of the lumbar spine in the X-ray film. Therefore, the degree of osteoporosis at each lumbar vertebrae position is determined based on the difference in the grayscale value of each lumbar pixel relative to the grayscale value of the normal lumbar spine in the X-ray film.

[0091] The difference between the grayscale value of the candidate pixel and the standard grayscale value of the normal vertebral area is used as the difference index corresponding to the candidate pixel. When the lumbar spine is abnormal, the color of the image of the vertebral area is darker, that is, the grayscale value of the pixel in the vertebral area image is larger; when the lumbar spine is normal, the grayscale value of the pixel in the vertebral area image is relatively small. Therefore, the difference obtained by subtracting the grayscale value of the candidate pixel from the standard grayscale value of the normal vertebral area is used as the difference index corresponding to the candidate pixel. In specific applications, the implementer sets the standard grayscale value of the normal vertebral area according to the specific situation.

[0092] By using the above method, the difference index corresponding to each pixel in the vertebral area can be obtained, such as Figure 2 As shown, the figure is a flow chart of the method for obtaining the impact degree value and the difference index.

[0093] Step S3, combining the impact degree value and the difference index to obtain a lumbar osteoporosis evaluation value; and obtaining an auxiliary detection result of lumbar osteoporosis of the person to be detected based on the lumbar osteoporosis evaluation value.

[0094] Since the pixel point influence degree value and the corresponding difference index obtained in this embodiment can reflect the lumbar osteoporosis condition of the person to be tested to a certain extent, in order to improve the accuracy of the analysis results, a preliminary evaluation of the lumbar osteoporosis condition of the person to be tested will be carried out in combination with the pixel point influence degree value and the corresponding difference index.

[0095] Specifically, the osteoporosis degree value of each vertebral area is obtained according to the influence degree value of bone overlap of each pixel point in each vertebral area and the corresponding difference index, the influence degree value is negatively correlated with the osteoporosis degree value, and the difference index is positively correlated with the osteoporosis degree value.

[0096] Among them, a positive correlation means that the dependent variable will increase as the independent variable increases, and the dependent variable will decrease as the independent variable decreases. It can be an additive relationship, a multiplicative relationship, etc., which is determined by actual application; a negative correlation means that the dependent variable will decrease as the independent variable increases, and the dependent variable will increase as the independent variable decreases. It can be a subtractive relationship, a division relationship, etc., which is determined by actual application.

[0097] In this embodiment, a specific calculation formula for the osteoporosis degree value is given, and the osteoporosis degree value of the kth vertebral area can be expressed as:

[0098] ;

[0099] in, represents the osteoporosis degree value of the kth vertebral area, represents the number of pixels in the kth vertebral area, Indicates the degree of influence of the t-th pixel in the k-th vertebral area on the bone overlap. Represents the standard gray value of the normal vertebral area, represents the gray value of the t-th pixel in the k-th vertebral area, represents the normalization function, Indicates the absolute value sign.

[0100] Characterizes the difference between the grayscale value of the t-th pixel in the k-th vertebral region and the standard grayscale value of the normal vertebral region. The larger the value, the greater the difference between the grayscale value of the t-th pixel in the k-th vertebral region and the standard grayscale value of the normal vertebral region. When the difference between the grayscale value of the pixel in the vertebral region and the standard grayscale value of the normal vertebral region is greater and the value of the degree of influence of bone overlap is smaller, the greater the osteoporosis degree value of the vertebral region.

[0101] By adopting the above method, the osteoporosis degree value of each vertebral area can be obtained. The maximum value of the osteoporosis degree values ​​of all vertebral areas is determined as the lumbar osteoporosis evaluation value.

[0102] The larger the lumbar osteoporosis evaluation value is, the more likely it is that the lumbar spine of the patient to be tested is abnormal. Therefore, when the lumbar osteoporosis evaluation value is greater than the preset osteoporosis evaluation threshold, it is determined that the lumbar spine of the person to be tested is suspected of osteoporosis, and the doctor should pay more attention to the patient's lumbar spine condition and conduct further examinations; when the lumbar osteoporosis evaluation value is less than or equal to the preset osteoporosis evaluation threshold, it is determined that the lumbar spine of the person to be tested is not suspected of osteoporosis. In this embodiment, the preset osteoporosis evaluation threshold is 0.7. In specific applications, the implementer can set it according to specific circumstances.

[0103] At this point, the method provided in this embodiment has been used to complete a preliminary judgment on the lumbar spine condition of the patient to be tested.

[0104] This embodiment first extracts the vertebral area in the lumbar spine X-ray image of the person to be tested in combination with the shape characteristics of the vertebral area. By analyzing the similarity of the grayscale distribution of different pixel points, it can effectively avoid the problem that the grayscale distribution of local positions with similar structures caused by lumbar curvature is inconsistent, thereby affecting the accuracy of the analysis result. Under normal circumstances, each vertebral structure on the lumbar spine should have similar characteristics. This embodiment uses the regularity of the position distribution between pixels with similar grayscale distribution to determine whether the similar performance of these similar pixels is caused by the similarity of the lumbar vertebral structure, and obtains the difference index corresponding to each pixel in the vertebral area, and then evaluates the degree of influence of bone overlap on each pixel point, and obtains the value of the degree of influence of bone overlap on each pixel point. Further, combined with the value of the degree of influence of bone overlap on the pixel point and the corresponding difference index, a preliminary judgment is made on whether the lumbar spine of the person to be tested has suspected osteoporosis. The doctor will make a further judgment based on the preliminary judgment, thereby improving the accuracy of the judgment result of suspected osteoporosis and effectively ensuring the detection result of lumbar osteoporosis of the person to be tested.

[0105] Example of an auxiliary detection system for lumbar osteoporosis based on X-ray film:

[0106] See also Figure 4 , which shows a structural block diagram of an X-ray-based lumbar osteoporosis auxiliary detection system provided by an embodiment of the present invention. The system may include an image acquisition module, a calculation module and a judgment module.

[0107] The image acquisition module is used to obtain the lumbar spine X-ray image of the person to be tested;

[0108] A calculation module is used to determine the vertebral area in combination with the shape distribution of the edge line in the lumbar X-ray image; screen similar pixels corresponding to each pixel in the vertebral area according to the difference between the local grayscale distribution of each pixel and other pixels in the vertebral area; obtain the value of the degree of influence of the bone overlap on each pixel based on the difference in the position distribution of each pixel and its similar pixels; obtain the difference index corresponding to each pixel in the vertebral area according to the difference between the standard grayscale of each pixel in the vertebral area and the normal vertebral area;

[0109] The judgment module is used to obtain a lumbar osteoporosis evaluation value by combining the impact degree value and the difference index; and obtain an auxiliary detection result of lumbar osteoporosis of the person to be detected based on the lumbar osteoporosis evaluation value.

[0110] It should be understood that Figure 4 The structural block diagram of the lumbar osteoporosis auxiliary detection system based on X-ray films and its modules shown can be implemented in various ways. For example, in some embodiments, the system and its modules can be implemented by hardware, software, or a combination of software and hardware. Among them, the hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated design hardware. Those skilled in the art can understand that the above-mentioned method and system can be implemented using computer executable instructions and / or included in a processor control code, for example, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. Such code is provided on the carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The system and its modules of this specification can not only be implemented by hardware circuits such as ultra-large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., but can also be implemented by software executed by various types of processors, and can also be implemented by a combination of the above-mentioned hardware circuits and software (for example, firmware).

[0111] For more details about the above modules, please refer to other places in this manual (for example Figure 1 , Figure 2 , Figure 3 part and its related description), which will not be repeated here.

[0112] In other embodiments, a device is also provided, including a memory and a processor. The memory is used to store executable program code, and the processor is used to call and run the executable program code from the memory, so that the device executes the above-mentioned X-ray-based lumbar osteoporosis auxiliary detection method. The device can be specifically a chip, a component or a module, and the chip may include a connected processor and a memory; wherein the memory is used to store instructions, and when the processor calls and executes the instructions, the chip can execute the X-ray-based lumbar osteoporosis auxiliary detection method provided in the above embodiment.

[0113] In other embodiments, a computer program product is also provided. When the computer program product is run on a computer, the computer executes the above-mentioned related steps to implement the X-ray-based lumbar osteoporosis auxiliary detection method provided in the above embodiment.

[0114] In other embodiments, a computer-readable storage medium is also provided, in which a computer program code is stored. When the computer program code is executed on a computer, the computer executes the above-mentioned related method steps to implement the X-ray-based lumbar osteoporosis auxiliary detection method provided in the above embodiment.

[0115] Among them, the provided systems, devices, computer program products, and computer-readable storage media are all used to execute the corresponding methods provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding methods provided above, and will not be repeated here.

[0116] It should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An X-ray-based auxiliary detection method for lumbar osteoporosis, characterized in that: The method comprises the following steps: Obtaining a lumbar spine X-ray image of the person to be tested; The vertebral area is determined by combining the shape distribution of the edge line in the lumbar X-ray image; the similar pixel points corresponding to each pixel point in the vertebral area are screened according to the difference between the local grayscale distribution of each pixel point and other pixel points in the vertebral area; the influence degree value of each pixel point affected by the bone overlap is obtained based on the difference in the position distribution of each pixel point and its similar pixel points; the difference index corresponding to each pixel point in the vertebral area is obtained according to the difference between the standard grayscale of each pixel point in the vertebral area and the normal vertebral area; Combining the impact degree value and the difference index, obtaining a lumbar vertebra osteoporosis evaluation value; obtaining an auxiliary detection result of lumbar vertebra osteoporosis of the person to be detected based on the lumbar vertebra osteoporosis evaluation value; The method of screening similar pixels corresponding to each pixel in the vertebral region according to the difference between the local grayscale distribution of each pixel in the vertebral region and other pixels includes: According to the difference in grayscale values ​​between the candidate pixel and the pixel points at corresponding positions in the neighborhood of each other pixel point in the vertebral region, a similarity index between the candidate pixel point and each other pixel point in the vertebral region is obtained; the candidate pixel point is any pixel point in the vertebral region; Based on the similarity index between each pixel point in the vertebral body region and each other pixel point in the vertebral body region, similar pixel points corresponding to each pixel point in the vertebral body region are screened; The method of obtaining the influence degree value of each pixel point caused by the overlap of bones based on the difference in the position distribution of each pixel point and its similar pixel points includes: According to the position coordinates of the candidate pixel point and its corresponding similar pixel point in the lumbar spine X-ray image, the fitting values ​​corresponding to the candidate pixel point and its corresponding similar pixel point are obtained; According to the difference between the horizontal coordinates of the candidate pixel point and its corresponding similar pixel point and the corresponding fitting value, the influence degree value of the candidate pixel point caused by the bone overlap is obtained, and the difference between the horizontal coordinate and the corresponding fitting value is negatively correlated with the influence degree value; The step of obtaining the fitting values ​​corresponding to the candidate pixel points and the similar pixel points corresponding to the candidate pixel points according to the position coordinates of the candidate pixel points and the similar pixel points corresponding to the candidate pixel points in the lumbar spine X-ray image includes: The candidate pixel and all its corresponding similar pixels are recorded as the first pixel; Obtaining position coordinates of all first pixel points in the lumbar spine X-ray image, and exchanging the abscissa and ordinate in the position coordinates of each first pixel point to obtain corresponding new position coordinates; Based on the new position coordinates of all the first pixel points, curve fitting is performed on all the first pixel points to obtain a fitting curve corresponding to the candidate pixel point; The ordinate on the fitting curve is used as the fitting value corresponding to the first pixel point.

2. The X-ray-based auxiliary detection method for lumbar osteoporosis according to claim 1, characterized in that: The step of determining the vertebral area by combining the shape distribution of the edge lines in the lumbar spine X-ray image includes: Divide each edge line in the lumbar spine X-ray image into several sub-segments; Determine the possible value of each sub-line segment being an arc shape according to the slope difference between adjacent pixel points on each sub-line segment, wherein the slope difference is positively correlated with the possible value of the arc shape; determine the sub-line segment corresponding to the possible value of the arc shape being greater than a preset first possible threshold as a suspected vertebral foramen edge line; Determine the possible value of each suspected vertebral foramen edge line as a vertebral foramen edge line according to the overall distribution of the slope of the pixel points on each suspected vertebral foramen edge line; determine the suspected vertebral foramen edge line corresponding to the possible value of the vertebral foramen edge line greater than a preset second possible threshold as the vertebral foramen edge line; The vertebral body area was obtained by connecting the edge lines of all vertebral foramina.

3. The X-ray-based auxiliary detection method for lumbar osteoporosis according to claim 1, characterized in that: The filtering of similar pixels corresponding to each pixel in the vertebral region based on a similarity index between each pixel in the vertebral region and each other pixel in the vertebral region includes: For a candidate pixel point, a pixel point whose similarity index with the candidate pixel point is greater than a preset similarity threshold is taken as a similar pixel point corresponding to the candidate pixel point.

4. The X-ray-based auxiliary detection method for lumbar osteoporosis according to claim 1, characterized in that: The step of obtaining a difference index corresponding to each pixel in the vertebral region according to the difference between each pixel in the vertebral region and the standard grayscale of a normal vertebral region includes: The difference between the gray value of the candidate pixel and the standard gray value of the normal vertebral area is used as the difference index corresponding to the candidate pixel.

5. The X-ray-based auxiliary detection method for lumbar osteoporosis according to claim 1, characterized in that: The combining the impact degree value and the difference index to obtain the lumbar osteoporosis evaluation value includes: Obtaining the osteoporosis degree value of each vertebral area according to the influence degree value of the bone overlap of each pixel point in each vertebral area and the corresponding difference index, wherein the influence degree value is negatively correlated with the osteoporosis degree value, and the difference index is positively correlated with the osteoporosis degree value; The maximum value of the osteoporosis degree values ​​of all vertebral areas is determined as the lumbar osteoporosis evaluation value.

6. The X-ray-based auxiliary detection method for lumbar osteoporosis according to claim 1, characterized in that: The auxiliary detection result of lumbar osteoporosis of the person to be detected is obtained based on the lumbar osteoporosis evaluation value, including: When the lumbar vertebra osteoporosis evaluation value is greater than the preset osteoporosis evaluation threshold, it is determined that the lumbar vertebra of the person to be tested has suspected osteoporosis; When the lumbar osteoporosis evaluation value is less than or equal to a preset osteoporosis evaluation threshold, it is determined that the lumbar spine of the person to be tested does not have suspected osteoporosis.

7. An X-ray-based lumbar osteoporosis auxiliary detection system, characterized in that: The system is used to implement the method according to any one of claims 1 to 6, and the system comprises: An image acquisition module is used to obtain a lumbar spine X-ray image of the person to be tested; A calculation module is used to determine the vertebral area in combination with the shape distribution of the edge line in the lumbar X-ray image; screen similar pixels corresponding to each pixel in the vertebral area according to the difference between the local grayscale distribution of each pixel and other pixels in the vertebral area; obtain the value of the degree of influence of the bone overlap on each pixel based on the difference in the position distribution of each pixel and its similar pixels; obtain the difference index corresponding to each pixel in the vertebral area according to the difference between the standard grayscale of each pixel in the vertebral area and the normal vertebral area; The judgment module is used to obtain a lumbar osteoporosis evaluation value by combining the impact degree value and the difference index; and obtain an auxiliary detection result of lumbar osteoporosis of the person to be detected based on the lumbar osteoporosis evaluation value.

Images