An intelligent positioning and annotation method for anorectal lesion areas

Abstract

The present invention relates to the technical field of lesion area localization in perianal MRI images, and specifically relates to an intelligent localization and annotation method for anorectal lesion areas. The present invention analyzes and obtains high-signal areas in perianal MRI images; according to the gray-scale differences in different areas, as well as the number of pixel points and gray-scale distribution in the areas, the perianal characterization degree is obtained, and then the initial areas are screened out; according to the position distribution and gradient distribution of pixel points on the texture curve in the initial areas, the direction extension coefficient is obtained; combining the gray-scale difference and distance difference between each pixel point in the area and its surroundings, the anorectal lesion localization coefficient is obtained; and then the anorectal lesion area is localized. The present invention can accurately identify the anorectal lesion areas in perianal MRI images, enabling relevant personnel to better judge the situation of anorectal lesions, thereby reducing the probability of misdiagnosis.

Description

Technical Field

[0001] The present invention relates to the technical field of locating pathological regions of anorectal images, and particularly to an intelligent positioning and annotation method for anorectal pathological regions. Background Art

[0002] MRI images are of great value in the examination of anal fistulas and have become the gold standard for preoperative evaluation in clinical practice. Their high soft tissue resolution can clearly show the course of the fistula tract, the location of the internal opening, and the relationship with the anal sphincter, helping doctors to accurately classify and plan the surgical plan. However, the lesions of anal fistulas are complex, the fistula tract may be tortuous and accompanied by branches, and traditional MRI examinations rely on manual interpretation, which has subjectivity and the risk of missed diagnosis. Therefore, it is particularly important to develop a positioning and annotation system. Through image processing technology, it can automatically identify the pathological region, accurately locate the fistula tract and the internal opening and perform annotation, which can not only improve the diagnostic efficiency, but also reduce human error, providing a more reliable basis for clinical treatment, especially having significant advantages in the diagnosis of complex anal fistulas and recurrent anal fistulas.

[0003] MRI images are widely used in the examination of perianal diseases. For example, in the case of anal fistulas, especially complex multi-branched fistula tracts, it is necessary to manually analyze the images during the diagnosis process, identify the high-signal regions and combine clinical experience to judge the path, morphology and fibrosis characteristics of the fistula tract. And in MRI images, the fistula tract usually shows as high-signal regions, and these regions may be similar to other structures, such as cysts or blood vessels, so the impact on subjective diagnosis will reduce the diagnostic efficiency and increase the risk of misdiagnosis. Summary of the Invention

[0004] In order to solve the technical problem that in MRI images, the fistula tract usually shows as high-signal regions, and these regions may be similar to other structures, such as cysts or blood vessels, so it is easy for relevant personnel to make mistakes when judging the pathological region, increasing the risk of misdiagnosis, the purpose of the present invention is to provide an intelligent positioning and annotation method for anorectal pathological regions, and the specific technical solution adopted is as follows:

[0005] An intelligent positioning and annotation method for anorectal pathological regions, the method includes:

[0006] Collect all perianal MRI images of the patient;

[0007] According to the gray-scale distribution of pixel points in each perianal MRI image, all high-signal regions in the perianal MRI image are obtained; the edge pixel points of each high-signal region are obtained; any one high-signal region is selected as the reference high-signal region; according to the number of pixel points and gray-scale distribution of the reference high-signal region, and the gray-scale difference between each edge pixel point and the surrounding region, the perianal characterization degree of the reference high-signal region is obtained; all high-signal regions are screened according to the perianal characterization degree to obtain the initial high-signal regions; any one initial high-signal region is selected as the target region; edge detection is performed on the target region to obtain all texture curves within the target region; according to the position distribution and gradient distribution of each pixel point on the texture curve within the target region, the direction extension coefficient of the target region is obtained; according to the gray-scale difference and distance difference between each pixel point within the target region and other pixel points within the preset neighborhood, and the direction extension coefficient of the target region, the anorectal lesion localization coefficient of the target region is obtained.

[0008] Locate the anorectal lesion area according to the anorectal lesion localization coefficient of each initial high-signal region.

[0009] Further, the method for obtaining the high-signal region includes:

[0010] The maximum inter-class variance method is used to segment the perianal MRI image, and all high-signal regions are screened out from the perianal MRI image.

[0011] Further, the method for obtaining the perianal characterization degree includes:

[0012] The region in the perianal MRI image other than the high-signal region is used as the low-signal region;

[0013] The perianal characterization degree is obtained according to the perianal characterization degree calculation formula, and the perianal characterization degree calculation formula is as follows:

[0014] ;

[0015] In the formula, represents the perianal characterization degree of the reference high-signal region; represents the number of pixel points within the minimum circumscribed circle of the reference high-signal region; represents the number of pixel points within the reference high-signal region; represents the standard deviation of the gray-scale values of the pixel points in the reference high-signal region; represents the th gray-scale value of the pixel points in the reference high-signal region; represents the number of edge pixel points of the reference high-signal region; represents the th gray-scale value of the edge pixel points of the reference high-signal region; represents the number of pixel points in the low-signal region; The gray value of the th pixel point representing the low-signal area.

[0016] Furthermore, the method for obtaining the initial high-signal area includes:

[0017] Taking the high-signal area with the perianal characterization degree greater than the preset first threshold as the initial high-signal area.

[0018] Furthermore, the method for obtaining the direction extension coefficient includes:

[0019] Establishing a Cartesian coordinate system with the lower left corner of the perianal MRI image as the origin, and selecting the midline of the perianal MRI image perpendicular to the x-axis as the reference line;

[0020] Calculating the shortest distance between each pixel point in the texture curve within the target area and the reference line as the first distance;

[0021] Obtaining the direction extension coefficient according to the direction extension coefficient calculation formula, and the direction extension coefficient calculation formula is as follows:

[0022] ;

[0023] In the formula, represents the direction extension coefficient of the target area; represents the number of texture curves within the target area; represents the th number of pixel points within the th texture curve within the target area; represents the th gradient of the th pixel point within the th texture curve within the target area; represents the th gradient of the th pixel point within the th texture curve within the target area;

[0024] Furthermore, the method for obtaining the anorectal lesion localization coefficient includes:

[0025] Obtaining the anorectal lesion localization coefficient according to the anorectal lesion localization coefficient calculation formula, and the anorectal lesion localization coefficient calculation formula is as follows:

[0026] ;

[0027] In the formula, represents the anorectal lesion localization coefficient of the target area; represents the direction extension coefficient of the target area; represents the number of pixel points in the target area; represents the distance between the th pixel point in the target area and the nearest pixel point with the same gray value; represents the number of other pixel points in the preset neighborhood of the th pixel point in the target area; represents the gray value of the th pixel point in the target area; represents the gray value of the th other pixel point in the preset neighborhood of the

[0028] Further, the anorectal lesion area is located according to the anorectal lesion location coefficient of each initial high-signal area, including:

[0029] Mark each initial high-signal area in each perianal MRI image with an anorectal lesion location coefficient greater than a preset second threshold, and divide the minimum circumscribed rectangle within the marked initial high-signal area to obtain all suspected lesion areas in each perianal MRI image;

[0030] Overlap the suspected lesion areas located at the same human position in all perianal MRI images, and use the suspected lesion area with the most overlap times as the anorectal lesion area.

[0031] An intelligent positioning and annotation system for anorectal lesion areas, the system includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned intelligent positioning and annotation method for anorectal lesion areas are implemented.

[0032] A computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the steps of the above-mentioned intelligent positioning and annotation method for anorectal lesion areas are implemented.

[0033] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned intelligent positioning and annotation method for anorectal lesion areas are implemented.

[0034] The present invention has the following beneficial effects:

[0035] In order to examine the perianal area of a patient, all perianal MRI images of the patient are first obtained; since anal fistula diseases exhibit high-signal characteristics in perianal MRI images, all high-signal regions in the perianal MRI images are located. However, there are other structural regions with relatively high gray-scale manifestations near the perianal area, so the manifestation characteristics of the fistula disease are analyzed to obtain the perianal characterization degree of the reference high-signal region; since there are significant differences in the edge characteristics between other tissues with high-signal characteristics and the fistula region, edge detection is first performed on each initial high-signal region to obtain all texture curves within each initial high-signal region, and the direction extension coefficient of the texture curve is used to judge which tissues the initial high-signal region specifically belongs to, obtaining the anorectal lesion localization coefficient of each high-signal region; the anorectal lesion region is located according to the anorectal lesion localization coefficient of each initial high-signal region. The present invention can accurately identify the anorectal lesion region in perianal MRI images, enabling relevant personnel to better judge the situation of anorectal lesions, thereby reducing the probability of misdiagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] 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 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, other drawings can be obtained based on these drawings without creative efforts.

[0037] Figure 1 It is a flowchart of a method for intelligent localization and annotation of anorectal lesion regions provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0038] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following will, in conjunction with the drawings and preferred embodiments, detail the specific implementation manner, structure, features, and effects of an intelligent localization and annotation method for anorectal lesion regions proposed 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.

[0039] 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.

[0040] The following will specifically describe the specific solution of an intelligent localization and annotation method for anorectal lesion regions provided by the present invention in conjunction with the drawings.

[0041] Please refer toFigure 1 , which shows an intelligent positioning and annotation method for anorectal lesion areas provided by an embodiment of the present invention. The method includes:

[0042] Step S1: Collect all perianal MRI images of the patient.

[0043] The embodiment of the present invention is mainly applied to the scenario of perianal disease examination of patients. In actual situations, patients usually need to lie supine in the scanner, and the scanner uses a body coil for imaging. During the scanning process, T2WI sequence scanning is performed on all human areas of the patient from the lower edge of the pubic symphysis to the tip of the coccyx. At this time, a number of perianal MRI images are collected by the scanner.

[0044] It should be noted that the method for collecting perianal MRI images is a well-known technical means to those skilled in the art and will not be limited and elaborated here.

[0045] In an embodiment of the present invention, the collected perianal MRI images are denoised to minimize the noise information in the images while retaining the details and edge information in the perianal MRI images, and finally the perianal MRI images required for subsequent operations are obtained. In an embodiment of the present invention, the wavelet transform method is used for denoising, and the denoising method is not unique. It should be noted that the wavelet transform method is a well-known technical means to those skilled in the art and will not be limited and elaborated here.

[0046] Step S2: Obtain all high-signal regions in the perianal MRI image according to the gray-scale distribution of pixel points in each perianal MRI image; obtain the edge pixel points of each high-signal region; arbitrarily select a high-signal region as a reference high-signal region; obtain the perianal characterization degree of the reference high-signal region according to the number and gray-scale distribution of pixel points in the reference high-signal region, and the gray-scale difference between each edge pixel point and the surrounding region; screen all high-signal regions according to the perianal characterization degree to obtain initial high-signal regions; arbitrarily select an initial high-signal region as a target region; perform edge detection on the target region to obtain all texture curves within the target region; obtain the direction extension coefficient of the target region according to the position distribution and gradient distribution of each pixel point on the texture curve within the target region; obtain the anorectal lesion localization coefficient of the target region according to the gray-scale difference and distance difference between each pixel point within the target region and other pixel points within the preset neighborhood, and the direction extension coefficient of the target region.

[0047] In reality, the gray-scale manifestation of anal fistula diseases in perianal MRI images is relatively prominent, that is, they present high-signal characteristics in perianal MRI images. Therefore, in the embodiments of the present invention, all high-signal regions in perianal MRI images are located. However, there will be other structural regions with relatively high gray-scale manifestations near the perianal region. For example, tissue edema may occur around the fistula due to inflammatory reactions, and the edema region has a high gray-scale in perianal MRI images. Therefore, in the embodiments of the present invention, the manifestation characteristics of fistula diseases are analyzed, and the perianal characterization degree of the reference high-signal region is obtained according to the number of pixel points and gray-scale distribution of the reference high-signal region, as well as the gray-scale difference between each edge pixel point and the surrounding region.

[0048] Preferably, in one embodiment of the present invention, the method for obtaining a high-signal region includes:

[0049] The present invention uses the OTSU threshold segmentation algorithm to select the optimal segmentation threshold for different regions in the perianal MRI image by maximizing the between-class variance, so as to segment the foreground as the high-signal region and the background as the low-signal region. A specific method provided by the embodiments of the present invention is as follows: calculate the gray-scale histogram of all pixel points in the perianal MRI image, extract the probability distribution of different gray levels therefrom, calculate the between-class variance under different thresholds by traversing all possible thresholds, and select the threshold corresponding to the maximum between-class variance as the segmentation threshold to complete the segmentation of the foreground and the background. It should be noted that the OTSU threshold segmentation algorithm is a well-known technical means in the art and will not be limited and described in detail here.

[0050] Preferably, in one embodiment of the present invention, the method for obtaining the perianal characterization degree includes:

[0051] Since the path of the fistula can usually be divided into straight, curved or spiral types, etc., the fistula presents obvious linear characteristics in morphological characterization. Therefore, the smallest circumscribed circle related to and closely related to the morphology is used to reflect the linear characteristics.

[0052] The perianal characterization degree is obtained according to the perianal characterization degree calculation formula, and the perianal characterization degree calculation formula is as follows:

[0053] ;

[0054] In the formula, represents the perianal characterization degree of the reference high-signal region; represents the number of pixel points within the smallest circumscribed circle of the reference high-signal region; represents the number of pixel points in the reference high-signal region; represents the standard deviation of the gray-scale values of the pixel points in the reference high-signal region; represents the th pixel point of the reference high-signal region; Indicates the number of edge pixels of the reference high-signal region; Indicates the gray value of the th edge pixel of the reference high-signal region; Indicates the number of pixels in the low-signal region; Indicates the

[0055] gray value of the th pixel in the low-signal region. In the perianal characterization degree calculation formula, the more the number of pixels of the minimum circumscribed circle of the reference high-signal region and the fewer the number of pixels of the reference high-signal region at this time, it indicates that the linear feature of the reference high-signal region is more prominent at this time, and the morphological feature of the reference high-signal region is more likely to be the fistula region, that is, the perianal characterization degree of the reference high-signal region is greater; the overall gray value in the reference high-signal region is greater, and the gray standard deviation

[0056] in the region is

[0057] smaller, the gray performance in the region is more stable, and at this time the reference high-signal region is more likely to be the fistula region, that is, the perianal characterization degree of the reference high-signal region is greater; since the fistula wall of the fistula region will show low-signal fibrous tissue, the gray performance of this tissue in the perianal MRI image is low, so the gray value of the edge pixels of the fistula region is low, so when the gray difference

[0058] between each edge pixel of the reference high-signal region and the internal pixels of the reference high-signal region is

[0059] greater, and the gray difference between each edge pixel of the reference high-signal region and the pixels of the low-signal region is

[0056] smaller, it indicates that the edge pixels more conform to the gray characteristics of fibrosis, and at this time the reference high-signal region is more likely to be the fistula region, that is, the perianal characterization degree of the reference high-signal region is greater.

[0056] Screen all high-signal regions according to the perianal characterization degree to obtain the initial high-signal region.

[0057] Preferably, in an embodiment of the present invention, the method for obtaining the initial high-signal region includes:

[0058] Taking the high-signal region with a perianal characterization degree greater than a preset first threshold as the initial high-signal region. In an embodiment of the present invention, the preset first threshold is set to 0.8. It should be noted that the preset first threshold can be set by oneself and is not limited here.

[0059] In the perianal MRI images, there are other tissues with high signal characteristics. However, there are significant differences in the edge characteristics between these tissues and the fistula area. For example, tissue edema may occur around the fistula due to the inflammatory reaction. However, different from the strip or tubular high signal of the fistula itself, the inflammatory edema will present a sheet-like or flocculent high signal due to the lack of a clear tube wall structure, and there is no clear lumen structure with a blurred boundary. The internal orifice of the fistula is mostly located at the midline posterior to the dentate line, and its growth path often extends along tissue spaces or the direction of least resistance. For example, the fistula of anal fistula often spreads along the sphincter space, that is, the potential space between the internal and external anal sphincters, thus forming a specific running direction. Pus or inflammatory secretions will flow along tissue spaces, forming the extension direction of the fistula. For example, if perianal abscess is not drained in time, the pus can spread to the deep or surrounding spaces, thus increasing the risk of forming complex multi-branched fistulas. Therefore, in the embodiments of the present invention, edge detection is first performed on each initial high signal region to obtain all texture curves within each initial high signal region, and the specific tissues to which the initial high signal region belongs are judged according to the trend of the texture curves.

[0060] Preferably, in one embodiment of the present invention, the method for obtaining the direction extension coefficient includes:

[0061] Since the internal orifice of the fistula is mostly located at the midline posterior to the dentate line, the position difference between the pixel points within each initial high signal region and the midline posterior to the dentate line is analyzed to judge whether the region belongs to the fistula region. Therefore, a Cartesian coordinate system is established with the lower left corner of the perianal MRI image as the origin, and the midline of the perianal MRI image perpendicular to the x-axis is selected as the reference line. At this time, the reference line is exactly the line where the midline posterior to the dentate line is located.

[0062] The shortest distance between the pixel points in each texture curve within the target region and the reference line is calculated as the first distance. It should be noted that calculating the shortest distance between each pixel point and the reference line is a well-known technical means for those skilled in the art and will not be elaborated here.

[0063] The direction extension coefficient is obtained according to the direction extension coefficient calculation formula. The direction extension coefficient calculation formula is as follows:

[0064] ;

[0065] In the formula, represents the direction extension coefficient of the target region; represents the number of texture curves within the target region; represents the th within the target region represents the th within the target region Gradient of a pixel Indicates the th texture curve within the target area and the gradient of the th pixel within it; Indicates the first distance of the

[0066] th pixel within the th texture curve within the target area. In the formula for calculating the direction extension coefficient, calculate the average gradient difference between every two adjacent pixels in each texture curve . If the average gradient difference is smaller, it indicates that the gradient direction of this texture curve is more consistent. At this time, analyze the within each texture curve. The smaller the , the larger the direction extension coefficient of the target area; if the number of pixels in the texture curve is larger, it indicates that the length of this texture curve is longer, and the larger the , the larger the direction extension coefficient of this texture curve within the target area. Analyze all texture curves. The larger the , the larger the direction extension coefficient of the target area; since the inner orifice of the fistula is mostly located around the reference line, calculate the average first distance between each pixel in each texture curve and the reference line

[0067] . The smaller the

[0068] , the greater the tendency of the target area to be a fistula area, that is, the larger the direction extension coefficient of the target area. Since the extensibility of the fistula is manifested as its gradual connection with the surrounding tissues, that is, the connectivity between adjacent parts of the tissues around the fistula area is relatively strong, localize the perianal MRI image by combining the connectivity with the direction extension coefficient of each initial high-signal area, and analyze the anorectal lesion localization coefficient of the high-signal area.

[0069] Preferably, in an embodiment of the present invention, the method for obtaining the anorectal lesion localization coefficient includes:

[0070] ;

[0071] In the formula, represents the anorectal lesion localization coefficient of the target area; represents the direction extension coefficient of the target area; represents the number of pixels in the target area; represents the distance between the ​denotes the number of other pixels within the preset neighborhood of the th pixel in the target area; denotes the gray value of the th pixel in the target area; denotes the gray value of the th other pixel within the preset neighborhood of the th pixel in the target area.

[0072] In the calculation formula of the anorectal lesion localization coefficient, if the gray difference between each pixel in the target area and other pixels within the preset neighborhood is smaller, and the distance between each pixel is smaller, it indicates that the connectivity between the pixels in the target area and the preset neighborhood is stronger. At this time, the target area is more likely to be a fistula area, that is, the anorectal lesion localization coefficient of the target area is larger; if the direction extension coefficient

[0073] of the target area is larger, it indicates that the target area is more likely to be a fistula area at this time, that is, the anorectal lesion localization coefficient of the target area is larger.

[0074] Step S3: Locate the anorectal lesion area according to the anorectal lesion localization coefficient of each initial high-signal area.

[0075] Preferably, in an embodiment of the present invention, the specific method includes:

[0076] Mark each initial high-signal area in each perianal MRI image with an anorectal lesion localization coefficient greater than a preset second threshold, and divide the minimum circumscribed rectangle within the marked initial high-signal area to obtain all suspected lesion areas in each perianal MRI image; in an embodiment of the present invention, the preset second threshold can be set to 0.9. It should be noted that the preset second threshold can be set by oneself and is not limited here.

[0077] Overlap the suspected lesion areas located at the same human position in all perianal MRI images, and use the suspected lesion area with the most overlap times as the anorectal lesion area.

[0078] Thus, the intelligent localization of the anorectal lesion area is completed.

[0079] In summary, all perianal MRI images of the patient are collected; all high-signal regions in the perianal MRI images are obtained according to the gray-scale distribution of pixel points in each perianal MRI image; the edge pixel points of each high-signal region are obtained; an arbitrary high-signal region is selected as the reference high-signal region; according to the number of pixel points and gray-scale distribution of the reference high-signal region, and the gray-scale difference between each edge pixel point and the surrounding region, the perianal characterization degree of the reference high-signal region is obtained; all high-signal regions are screened according to the perianal characterization degree to obtain the initial high-signal regions; an arbitrary initial high-signal region is selected as the target region; edge detection is performed on the target region to obtain all texture curves within the target region; according to the position distribution and gradient distribution of each pixel point on the texture curve within the target region, the direction extension coefficient of the target region is obtained; according to the gray-scale difference and distance difference between each pixel point within the target region and other pixel points within the preset neighborhood, and the direction extension coefficient of the target region, the anorectal lesion localization coefficient of the target region is obtained; the anorectal lesion region is located according to the anorectal lesion localization coefficient of each initial high-signal region.

[0080] The second object of an embodiment of the present invention is to provide an intelligent positioning and annotation system for anorectal lesion regions, which includes a memory, a processor, and a computer program, where the memory is used to store the corresponding computer program, the processor is used to run the corresponding computer program, and when the computer program runs in the processor, it can implement the method described in steps S1-S3.

[0081] The third object of an embodiment of the present invention is to provide a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the method described in steps S1-S3.

[0082] The fourth object of an embodiment of the present invention is to provide a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the method described in steps S1-S3.

[0083] It should be noted that the above sequence of embodiments of the present invention is only for description and does not represent the advantages or disadvantages 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.

[0084] Each embodiment in this specification is described in a progressive manner. The same or similar parts among the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. An intelligent positioning and marking method for anorectal lesion areas, characterized in that: The method comprises: All perianal MRI images of the patients were collected; According to the grayscale distribution of pixels in each of the perianal MRI images, all high-signal areas in the perianal MRI images are obtained; the edge pixels of each high-signal area are obtained; one high-signal area is selected as a reference high-signal area; according to the number of pixels and grayscale distribution in the reference high-signal area, and the grayscale difference between each edge pixel and the surrounding area, the perianal characterization degree of the reference high-signal area is obtained; according to the perianal characterization degree, all high-signal areas are screened to obtain an initial high-signal area; one initial high-signal area is selected as a target area; edge detection is performed on the target area to obtain all texture curves in the target area; according to the position distribution and gradient distribution of each pixel on the texture curve in the target area, the directional extension coefficient of the target area is obtained; according to the grayscale difference and distance difference between each pixel in the target area and other pixels in a preset neighborhood, as well as the directional extension coefficient of the target area, the anorectal lesion positioning coefficient of the target area is obtained; Positioning the anorectal lesion area according to the anorectal lesion positioning coefficient of each initial high signal area; The method for obtaining the anorectal lesion positioning coefficient includes: The anorectal lesion positioning coefficient is obtained according to the anorectal lesion positioning coefficient calculation formula, and the anorectal lesion positioning coefficient calculation formula is as follows: ； In the formula, Indicates the anorectal lesion localization coefficient of the target area; Indicates the directional extension coefficient of the target area; Indicates the number of pixels in the target area; Indicates the target area The distance between a pixel and the nearest pixel with the same gray value; Indicates the target area The number of other pixels in the preset neighborhood of a pixel; Indicates the target area The gray value of each pixel; Indicates the target area The preset neighborhood of pixels Grayscale values ​​of other pixels; The anorectal lesion area is located according to the anorectal lesion localization coefficient of each initial high signal area, including: Marking each initial high signal area in each of the perianal MRI images whose anorectal lesion localization coefficient is greater than a preset second threshold, and dividing the minimum circumscribed rectangle excluding the marked initial high signal area to obtain all suspected lesion areas in each perianal MRI image; All suspected lesion areas located at the same human body position in the perianal MRI images were overlapped, and the suspected lesion area with the most overlaps was taken as the anorectal lesion area.

2. The intelligent positioning and marking method for anorectal lesion areas according to claim 1, characterized in that: The method for acquiring the high signal area comprises: The maximum inter-class variance method is used to segment the perianal MRI image, and all high signal areas are screened out from the perianal MRI image.

3. The intelligent positioning and marking method for anorectal lesion area according to claim 1, characterized in that: The method for obtaining the anal perianal characterization degree comprises: The areas other than the high signal areas in the perianal MRI images were regarded as low signal areas; The anal perianal characterization degree is obtained according to the anal perianal characterization degree calculation formula, and the anal perianal characterization degree calculation formula is as follows: ； In the formula, It indicates the degree of perianal representation of the reference high signal area; Represents the number of pixels within the minimum circumscribed circle of the reference high signal area; Indicates the number of pixels in the reference high signal area; Represents the standard deviation of the grayscale values ​​of pixels in the reference high signal area; Indicates the reference high signal area The gray value of each pixel; Indicates the number of edge pixels in the reference high signal area; Indicates the reference high signal area Gray value of edge pixels; The number of pixels representing low signal areas; Indicates the low signal area The gray value of a pixel.

4. The intelligent positioning and marking method for anorectal lesion area according to claim 1, characterized in that: The method for acquiring the initial high signal area comprises: The high signal area with the perianal characterization degree greater than the preset first threshold is taken as the initial high signal area.

5. The intelligent positioning and marking method for anorectal lesion area according to claim 1, characterized in that: The method for obtaining the directional extension coefficient includes: A Cartesian coordinate system is established with the lower left corner of the perianal MRI image as the origin, and the image midline of the perianal MRI image perpendicular to the x-axis is selected as a reference straight line; Calculate the shortest distance between each pixel point in the texture curve in the target area and the reference straight line as the first distance; The directional extension coefficient is obtained according to the directional extension coefficient calculation formula, and the directional extension coefficient calculation formula is as follows: ； In the formula, Indicates the directional extension coefficient of the target area; Indicates the number of texture curves within the target area; Indicates the first The number of pixels in a texture curve; Indicates the first Texture curve The gradient of each pixel; Indicates the first Texture curve The gradient of each pixel; Indicates the first Texture curve The first distance of pixels.

6. An intelligent positioning and marking system for anorectal lesion areas, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method for intelligent positioning and marking of anorectal lesion areas as described in any one of claims 1 to 5 are implemented.

7. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method for intelligent positioning and marking of anorectal lesion areas as described in any one of claims 1 to 5 are implemented.

8. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method for intelligent positioning and marking of anorectal lesion areas as described in any one of claims 1 to 5 are implemented.

Images