Intelligent pelvic floor muscle state evaluation method based on medical imaging analysis

Abstract

The invention relates to the technical field of pelvic floor muscle assessment, in particular to an intelligent pelvic floor muscle state assessment method based on medical imaging analysis. According to the method, the muscle edge in the pelvic floor muscle ultrasonic image is divided into edge segments, and according to the bending degree difference and the distribution direction difference of the edge segments in the image in the corresponding edge segments in two adjacent pelvic floor muscle ultrasonic images and the bending degree difference of the corresponding edge segments relative to the muscle edge where the corresponding edge segments are located, the pelvic floor muscle ultrasonic image is obtained. The muscle change complexity is obtained, the preset step length is adjusted according to the clearness of the edge segments in the pelvic floor muscle ultrasonic image corresponding to the edge segments, and the optimized step length of the edge segments in the analysis image is determined; and the pelvic floor muscle ultrasonic image is sampled by using the ultrasonic sensor, and the pelvic floor muscle state of the patient is evaluated based on the obtained position change of the sampling point. The preset step length is adaptively adjusted according to the pelvic floor muscle edge change and the clarity, and the accuracy of pelvic floor muscle state evaluation is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of pelvic floor muscle assessment, and particularly relates to an intelligent assessment method for pelvic floor muscle status using medical imaging analysis. Background Art

[0002] Female pelvic floor dysfunction is a common disease among middle-aged and elderly women. In the pelvic floor support structure, the supporting role of the pelvic floor muscle group is crucial. Therefore, the assessment of pelvic floor muscle status has important clinical value for the diagnosis and treatment of female pelvic floor dysfunction.

[0003] When assessing the status of pelvic floor muscles, it is necessary to analyze muscle movement, track the displacement of pelvic floor muscles during contraction, and extract the displacement curve of feature points. During the acquisition of the displacement curve, the feature points are usually tracked using a fixed step size. However, due to the complex pelvic floor structure, the changes of different muscles during contraction are different, and the pelvic floor muscle status of different patients varies. If the step size is too large, the subtle changes of pelvic floor muscles may not be accurately captured. If the step size is too small, the examination efficiency will be reduced, resulting in errors in the assessment of pelvic floor muscle status and low efficiency. Summary of the Invention

[0004] In order to solve the technical problem that the fixed step size in the feature selection process leads to errors in the assessment of pelvic floor muscle status, the purpose of the present invention is to provide an intelligent assessment method for pelvic floor muscle status using medical imaging analysis. The specific technical solutions adopted are as follows: The present invention proposes an intelligent assessment method for pelvic floor muscle status using medical imaging analysis. The method includes: Obtaining the pelvic floor muscle ultrasound video of a patient; Performing edge detection on each frame of pelvic floor muscle ultrasound image to obtain muscle edges, and dividing the corresponding muscle edges into edge segments according to the clarity of pixel points on each muscle edge; Taking the first frame of pelvic floor muscle ultrasound image as the analysis image, and obtaining the muscle change complexity of each edge segment in the analysis image according to the difference in bending degree and distribution direction of the corresponding edge segments in each adjacent two frames of pelvic floor muscle ultrasound images, as well as the difference in bending degree of the corresponding edge segment relative to the muscle edge where it is located; Adjusting the preset step size according to the muscle complexity change degree of each edge segment in the analysis image and the clarity of the corresponding edge segment in the pelvic floor muscle ultrasound image, and determining the optimized step size of each edge segment in the analysis image; Evaluating the pelvic floor muscle status of the patient according to the position change of the sampling points obtained by sampling the pelvic floor muscle ultrasound image using the optimized step size.

[0005] Further, dividing the corresponding muscle edge into edge segments according to the clarity of the pixel points on each muscle edge includes: For each edge pixel point in the pelvic floor muscle ultrasound image, using the muscle edge where the edge pixel point is located, dividing the preset neighborhood of the edge pixel point into two local neighborhood regions; obtaining the gray concentration value of the local neighborhood region; According to the difference and gradient value of the gray concentration values of the two local neighborhood regions of the edge pixel point, obtaining the point clarity of the edge pixel point; Clustering the edge pixel points on each muscle edge based on the point clarity to obtain several clustering clusters; the edge pixel points on each edge segment are connected and within the same clustering cluster.

[0006] Further, obtaining the muscle change complexity of each edge segment in the analysis image includes: Obtaining the curvature of each edge segment; Arranging the corresponding edge segments of each edge segment in the analysis image in all pelvic floor muscle ultrasound images in time sequence to obtain the segment sequence of each edge segment in the analysis image; according to the difference in the curvature degree and the difference in the distribution direction between every two adjacent edge segments in the segment sequence, obtaining the motion deviation degree of each edge segment in the analysis image; Obtaining the curvature of the muscle edge where each edge segment is located; selecting the characteristic segment of each edge segment in the analysis image from the segment sequence, and according to the motion deviation degree of each edge segment in the analysis image and the difference between the characteristic segment of each edge segment and the curvature of the muscle edge where the characteristic segment is located, obtaining the muscle change complexity of each edge segment in the analysis image.

[0007] Further, obtaining the curvature of each edge segment includes: Performing linear fitting on the edge pixel points on each edge segment to obtain a distribution line; calculating the average distance from all edge pixel points on each edge segment to the distribution line as the overall spacing; According to the average curvature of the edge pixel points on each edge segment and the overall spacing, obtaining the curvature of the corresponding edge segment, and both the average curvature and the overall spacing are positively correlated with the curvature.

[0008] Further, obtaining the motion deviation degree of each edge segment in the analysis image includes: According to the absolute value of the difference between the angle between the distribution lines of each edge segment and its next edge segment in the segment sequence of each edge segment in the analysis image and the curvature, obtaining the local deviation degree of each edge segment in the segment sequence; the absolute value of the difference between the angle and the curvature is positively correlated with the local deviation degree; Take the sum of the local deviation degrees of the remaining edge segments except the last one in the sequence of segmenting each edge in the analysis image as the motion deviation degree of each edge segment in the analysis image.

[0009] Further, the determining the optimized step size for each edge segment in the analysis image includes: Take the sum of the point sharpness of the edge pixels on each edge segment as the edge sharpness; Determine the step size adjustment coefficient for the corresponding edge segment according to the muscle change complexity of each edge segment in the analysis image and the edge sharpness of its characteristic segment; there is a negative correlation between the edge sharpness and the step size adjustment coefficient, and there is a positive correlation between the muscle change complexity and the step size adjustment coefficient; Use the step size adjustment coefficient to weight the preset step size to obtain the optimized step size for each edge segment in the analysis image.

[0010] Further, the evaluating the pelvic floor muscle state of the patient according to the position change of the sampling points obtained by sampling the pelvic floor muscle ultrasound image using the optimized step size includes: Perform optical flow tracking on each edge pixel in the analysis image in all pelvic floor muscle ultrasound images to determine the optical flow matching point of each edge pixel, and arrange each edge pixel and its optical flow matching point in time sequence to obtain the tracking sequence of each edge pixel in the analysis image; Calculate the distances between the first pixel point in the tracking sequence and the remaining pixel points respectively, and take the maximum value of the distances as the feature screening index of each edge pixel in the analysis image; select the edge pixel corresponding to the maximum feature screening index on each edge segment in the analysis image as the feature point; Based on the optimized step size of the edge segment where each feature point in the analysis image is located, sample the pixel points in the tracking sequence of each feature point, and the obtained sampling points form the sampling sequence of each feature point; Establish a two-dimensional coordinate system with time as the horizontal axis and the displacement amount of the pixel point as the vertical axis; mark the distances between each pixel point in the sampling sequence and the first pixel point in the two-dimensional coordinate system to obtain the coordinate points of each pixel point; perform curve fitting on the coordinate points in the two-dimensional coordinate system to obtain the displacement curve of each feature point; Evaluate the pelvic floor muscle state of the patient based on the displacement curve of the feature points in the analysis image.

[0011] Further, the characteristic segment is the edge segment corresponding to the maximum curvature in the sequence of segmenting each edge in the analysis image.

[0012] Further, the muscle edge is an edge obtained by performing curve fitting on edge pixel points obtained by performing edge detection on each frame of pelvic floor muscle ultrasound image.

[0013] Further, the gray-scale concentration value is equal to the average gray scale of all pixel points in the local neighborhood area.

[0014] The present invention has the following beneficial effects: In the embodiment of the present invention, since the contraction strengths of different edge parts of the same pelvic floor muscle are different during the movement process, in order to more accurately evaluate the state of the pelvic floor muscle, the muscle edge is segmented based on the clarity of pixel points on the muscle edge to obtain edge segments; because the muscle edge changes to varying degrees during the movement process of the pelvic floor muscle and there are differences in the position changes of the muscle edge, resulting in changes in the curvature and overall direction of the muscle edge, considering the movement offset degree of the edge segments, and the difference between the edge segments and the curvature of the muscle edge where they are located reflects the abnormal degree of muscle tension or structure of the edge segments, which can reflect the complexity of muscle changes of the edge segments, comprehensively analyze to obtain the muscle change complexity of the edge segments; adaptively adjust the preset step size in combination with the muscle change complexity and clarity of the edge segments, and using the optimized step size for sampling can significantly improve the accuracy and fineness of capturing muscle changes during the movement process of the pelvic floor muscle, while reducing the error risk, providing more accurate and reliable data support for the state evaluation of the pelvic floor muscle, and effectively improving the accuracy of pelvic floor muscle state evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0016] Figure 1 It is a flowchart of the steps of an intelligent evaluation method for pelvic floor muscle state using medical imaging analysis provided by an embodiment of the present invention; Figure 2 It is a flowchart of the steps of a method for obtaining the muscle change complexity of edge segments provided by an embodiment of the present invention; Figure 3 It is a schematic diagram of a computer device of an intelligent device for pelvic floor muscle state using medical imaging analysis provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] To further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details a method for intelligent assessment of pelvic floor muscle status using medical imaging analysis according to the present invention, including its specific implementation manner, structure, features, and effects. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0019] The following specifically describes the specific solution of a method for intelligent assessment of pelvic floor muscle status using medical imaging analysis provided by the present invention in conjunction with the accompanying drawings.

[0020] Example 1: The present invention proposes a method for intelligent assessment of pelvic floor muscle status using medical imaging analysis. Please refer to Figure 1 , which shows the flowchart of the steps of a method for intelligent assessment of pelvic floor muscle status using medical imaging analysis provided by an embodiment of the present invention. The method includes: Step S1: Obtain the pelvic floor muscle ultrasound video of the patient.

[0021] Specifically, before the examination, ask the female patient to empty the bladder and bowel. The doctor guides the patient to adopt a suitable position, such as the supine position or the lateral position, for easy probe placement and observation. During the examination, first, use a transperineal ultrasound probe and place it on the perineum of the patient to ensure that the probe is in close contact with the skin. Then, by adjusting the probe angle and position, make the ultrasonic beam perpendicular to the pelvic floor muscles to obtain clear images. Finally, start the ultrasound device and continuously collect the pelvic floor muscle ultrasound video while the patient performs pelvic floor muscle contraction and relaxation movements.

[0022] It should be noted that the pelvic floor muscle ultrasound video is composed of multiple frames of pelvic floor muscle ultrasound images; the pelvic floor muscle ultrasound video only contains one pelvic floor muscle contraction and relaxation movement.

[0023] Step S2: Perform edge detection on each frame of pelvic floor muscle ultrasound image to obtain the muscle edge, and divide the corresponding muscle edge into edge segments according to the clarity of the pixel points on each muscle edge.

[0024] Since the edge shape and position changes of the pelvic floor muscles can reflect the pelvic floor function, the pelvic floor muscle edge is an important indicator for evaluating the pelvic floor muscle function. The muscle edge is obtained by curve fitting of the edge pixel points obtained by performing edge detection on each frame of pelvic floor muscle ultrasound image.

[0025] In this embodiment, the Canny operator is selected to perform edge detection on the pelvic floor muscle ultrasound image, and the least squares method is used to perform curve fitting on the edge pixel points in the pelvic floor muscle ultrasound image. Among them, the Canny operator and the least squares method are well-known technologies to those skilled in the art and will not be elaborated here.

[0026] During the contraction and relaxation movements of the patient's pelvic floor muscles, the contraction strengths of different parts of the muscles are different, resulting in differences in the clarity of different edge parts of the same muscle edge in the ultrasound image. In order to more accurately evaluate the state of the pelvic floor muscles, the muscle edge is segmented based on the clarity of the pixel points on the muscle edge, and edge segmentation is obtained.

[0027] Preferably, in some possible implementation manners of the embodiment of the present invention, the division method of the edge segmentation includes: for each edge pixel point in the pelvic floor muscle ultrasound image, using the muscle edge where the edge pixel point is located, dividing the preset neighborhood of the edge pixel point into two local neighborhood regions; obtaining the gray concentration value of the local neighborhood region; obtaining the point clarity of the edge pixel point according to the difference between the gray concentration values of the two local neighborhood regions of the edge pixel point and the gradient value; clustering the edge pixel points on each muscle edge based on the point clarity to obtain a plurality of clustering clusters; the edge pixel points on each edge segment are connected and in the same clustering cluster.

[0028] The gray concentration value reflects the overall gray level of the local neighborhood region; if the overall gray difference between the two local neighborhood regions of the edge pixel point, that is, the difference between the gray concentration values, is greater, it indicates that the edge pixel point is clearer in its preset neighborhood; if the gradient value of the edge pixel point is greater, it indicates that the gray difference between the edge pixel point and its surrounding pixels is greater, and then the edge pixel point is clearer and the point clarity is greater. Therefore, both the absolute value of the difference between the gray concentration values of the two local neighborhood regions of the edge pixel point and the gradient value are positively correlated with the point clarity. In the embodiment of the present invention, the product of the gradient value of the edge pixel point and the absolute value of the difference between the gray concentration values of its two local neighborhood regions is used as the point clarity of the edge pixel point.

[0029] In the embodiment of the present invention, the correlation relationship between the gradient value of the edge pixel point and the absolute value of the difference between the gray concentration values of its two local neighborhood regions can also be constructed through other basic mathematical operations, such as the sum value, which is not limited and elaborated here.

[0030] It should be noted that the local neighborhood region of the edge pixel point does not include the muscle edge where the edge pixel point is located.

[0031] In one implementation manner of the embodiment of the present invention, the K-means clustering algorithm is selected to cluster the edge pixel points on the edge segment, where K is equal to 3, and the implementer can set it according to the specific situation.

[0032] In an implementation manner of the embodiment of the present invention, the preset neighborhood is an eight-neighborhood.

[0033] Step S3: Denote the first frame of pelvic floor muscle ultrasound image as the analysis image, and obtain the muscle change complexity of each edge segment in the analysis image according to the differences in the bending degree and the distribution direction of the corresponding edge segments in each adjacent two frames of pelvic floor muscle ultrasound images, as well as the difference in the bending degree of the corresponding edge segment relative to the muscle edge where it is located.

[0034] During the movement of the pelvic floor muscles, there are varying degrees of changes in the muscle edges and differences in the position changes of the muscle edges, resulting in changes in the bending degree and the overall direction of the muscle edges; by analyzing the differences in the bending degree and the distribution direction of the corresponding edge segments in adjacent pelvic floor muscle ultrasound images, the movement offset degree of the edge segments is considered; the difference in the bending degree between the edge segment and the muscle edge where it is located reflects the abnormality degree of the muscle tension or structure of the edge segment, and further reflects the complexity of the muscle changes where the edge segment is located, so as to obtain the muscle change complexity of the edge segment.

[0035] The method for obtaining the corresponding edge segment of an edge segment in the pelvic floor muscle ultrasound image in the analysis image is as follows: Arbitrarily select an edge segment of the analysis image as the example segment, perform optical flow tracking on the edge pixel points on the example segment, determine the optical flow matching points of each edge pixel point on the example segment in the second frame of pelvic floor muscle ultrasound image, count the total number of optical flow matching points on each edge segment in the second frame of pelvic floor muscle ultrasound image, and select the edge segment corresponding to the largest total number as the corresponding edge segment of the example segment in the second frame of pelvic floor muscle ultrasound image. The method for obtaining the corresponding edge segment of each edge segment in the analysis image in all pelvic floor muscle ultrasound images is the same as the method for obtaining the corresponding edge segment of the example segment in the second frame of pelvic floor muscle ultrasound.

[0036] It should be noted that in this embodiment, the Lucas-Kanade method is used to perform optical flow tracking on the pixel points. The corresponding edge segment of the example segment in the first frame of pelvic floor muscle ultrasound image is the example segment itself.

[0037] Please refer to Figure 2 , which shows the flowchart of the steps of a method for obtaining the muscle change complexity of an edge segment provided by an embodiment of the present invention. The method includes: Step S310: Obtain the bending degree of each edge segment.

[0038] In some possible implementation manners of the embodiments of the present invention, the method for obtaining the curvature includes: performing linear fitting on the edge pixel points on each edge segment to obtain a distribution line; calculating the average distance from all the edge pixel points on each edge segment to the distribution line as the overall spacing; and obtaining the curvature of the corresponding edge segment according to the average curvature of the edge pixel points on each edge segment and the overall spacing.

[0039] In this embodiment, the curvature of the edge segment is measured by the distance from the edge pixel points on the edge segment to its distribution line and the curvature of the edge pixel points. If the distance from the edge pixel points on the edge segment to its distribution line is larger and the curvature is larger, it indicates that the curvature of the edge segment is more obvious, and then the curvature of the edge segment is larger. Therefore, both the average curvature and the overall spacing are positively correlated with the curvature. In the embodiments of the present invention, the product of the average curvature of the edge pixel points on each edge segment and the overall spacing is used as the curvature of the corresponding edge segment.

[0040] In the embodiments of the present invention, a correlation relationship between the average curvature of the edge pixel points, the overall spacing, and the curvature can also be constructed through other basic mathematical operations, such as the sum value, which is not limited and elaborated herein.

[0041] In the embodiments of the present invention, the least squares method is selected to perform linear fitting on the edge pixel points on the edge segment.

[0042] Step S320: Arrange the corresponding edge segments of each edge segment in the analysis image in all pelvic floor muscle ultrasound images in chronological order to obtain a segment sequence of each edge segment in the analysis image; and obtain the motion deviation degree of each edge segment in the analysis image according to the difference in the curvature and the difference in the distribution direction between every two adjacent edge segments in the segment sequence.

[0043] In some possible implementation manners of the embodiments of the present invention, the method for obtaining the motion deviation degree includes: obtaining the local deviation degree of each edge segment in the segment sequence according to the absolute value of the difference between the angle between the distribution lines of each edge segment and its next edge segment and the curvature in the segment sequence of each edge segment in the analysis image; the absolute value of the difference between the angle and the curvature is positively correlated with the local deviation degree; and taking the cumulative sum of the local deviation degrees of the edge segments except the last edge segment in the segment sequence of each edge segment in the analysis image as the motion deviation degree of each edge segment in the analysis image.

[0044] In a specific implementation manner of the embodiments of the present invention, the motion deviation degree is expressed by the formula: ; Wherein, R is the motion deviation degree of each edge segment in the analyzed image; N is the total number of edge segments in the segment sequence of each edge segment in the analyzed image, that is, the total number of pelvic floor muscle ultrasound images in the pelvic floor muscle ultrasound video; is the included angle between the distribution lines of the nth and (n + 1)th edge segments in the segment sequence of each edge segment in the analyzed image; is the curvature of the nth edge segment in the segment sequence of each edge segment in the analyzed image; is the curvature of the (n + 1)th edge segment in the segment sequence of each edge segment in the analyzed image; is the local deviation degree of the nth edge segment in the segment sequence of each edge segment in the analyzed image; is the absolute value function.

[0045] It should be noted that the direction of the distribution line of the edge segment reflects the overall distribution direction of the edge segment; if and are larger, it indicates that the direction change of the corresponding edge segment of each edge segment in the analyzed image in the nth and (n + 1)th pelvic floor muscle ultrasound images is larger and the curvature change is more obvious, then the motion change of each edge segment in the analyzed image during the pelvic floor muscle contraction is more obvious and the motion deviation degree is larger.

[0046] Step S330: Obtain the curvature of the muscle edge where each edge segment is located; select the characteristic segment of each edge segment in the analyzed image from the segment sequence, and obtain the muscle change complexity of each edge segment in the analyzed image according to the motion deviation degree of each edge segment in the analyzed image and the difference between the characteristic segment of each edge segment and the curvature of the muscle edge where the characteristic segment is located.

[0047] The motion deviation degree reflects the motion change information of the edge segment during the examination, and the curvature reflects the bending information of the edge segment at a single moment. In order to make the motion change of the edge segment in the pelvic floor muscle more sensitive, the edge segment corresponding to the maximum curvature in the segment sequence of each edge segment in the analyzed image is selected as the characteristic segment.

[0048] The curvature of the pelvic floor muscle edge means the change of muscle tension, function or structure; if the difference between the edge segment and the curvature of the muscle edge where it is located is larger, the muscle edge is often affected by multiple factors together, indicating that the muscle tension or structure of this edge segment is more abnormal, then the muscle change of this edge segment is more complex. If the motion deviation degree is larger, it indicates that the change of each edge segment in the analyzed image during the pelvic floor muscle movement is more obvious, meaning that the muscle change of the edge segment is more complex. Therefore, both the motion deviation degree of each edge segment in the analyzed image and the difference between the characteristic segment of each edge segment and the curvature of the muscle edge of the characteristic segment are positively correlated with the muscle change complexity.

[0049] In a specific implementation manner of the embodiment of the present invention, the muscle change complexity is expressed by the formula: ; In the formula, W is the muscle change complexity of each edge segment in the analysis image; is the curvature of the feature segment of each edge segment in the analysis image; is the curvature of the muscle edge where the feature segment of each edge segment in the analysis image is located; is the motion deviation degree of each edge segment in the analysis image; is the absolute value function.

[0050] It should be noted that the method for obtaining the curvature of the muscle edge is the same as that of the curvature of the edge segment.

[0051] Step S4: Adjust the preset step size according to the muscle complex change degree of each edge segment in the analysis image and the clarity of the corresponding edge segment in the pelvic floor muscle ultrasound image, and determine the optimized step size of each edge segment in the analysis image.

[0052] The edge segment with a greater muscle change complexity changes more complexly during the pelvic floor muscle movement. In order to avoid losing important details or being unable to capture the fine dynamic changes of the muscle, a smaller step size needs to be used to track these complex changes to ensure that the muscle change details are accurately captured; the lower the clarity of the edge segment increases the risk of error. Using a smaller step size to track the muscle change enables the actual muscle boundary and change to be clearly identified and traced in the blurred area. Combining the muscle complex change degree and edge clarity of the edge segment to adjust the preset step size, and sampling based on the obtained optimized step size can better capture complex or blurred changes, thereby more accurately reflecting the dynamic activity of the pelvic floor muscle.

[0053] In some possible implementation manners of the embodiment of the present invention, the method for obtaining the optimized step size includes: taking the sum of the point clarities of the edge pixel points on each edge segment as the edge clarity; determining the step size adjustment coefficient of the corresponding edge segment according to the muscle change complexity of each edge segment in the analysis image and the edge clarity of its feature segment; the edge clarity and the step size adjustment coefficient are negatively correlated, and the muscle change complexity and the step size adjustment coefficient are positively correlated; using the step size adjustment coefficient to weight the preset step size to obtain the optimized step size of each edge segment in the analysis image.

[0054] In a specific implementation manner of the embodiment of the present invention, the optimized step size is expressed by the formula: ; In the formula, For analyzing the optimized step size of each edge segment in the image; CL is the edge clarity of each edge segment in the analyzed image; W is the muscle change complexity of each edge segment in the analyzed image; It is the step size adjustment coefficient for each edge segment in the analyzed image; B is the preset step size; It is a preset positive number, taking the empirical value 0.1, and its function is to prevent the fraction from being meaningless due to the denominator being zero; Norm is the normalization function. It should be noted that the value range of the optimized step size is .

[0055] In an implementation manner of the embodiment of the present invention, the preset step size is set to 5% of the total number of pixel points in the height direction of the pelvic floor muscle ultrasound image.

[0056] Step S5: Evaluate the pelvic floor muscle state of the patient according to the position change of the sampling points obtained by sampling the pelvic floor muscle ultrasound image using the optimized step size.

[0057] Sampling using the optimized step size can significantly improve the accuracy and fineness of capturing muscle changes during the movement of the pelvic floor muscles, while reducing the risk of errors, enabling the displacement curve of the feature points to accurately capture the subtle changes of the pelvic floor muscles, and providing more accurate and reliable data support for the state evaluation of the pelvic floor muscles.

[0058] In some possible implementation manners of the embodiment of the present invention, the method for obtaining the displacement curve includes: performing optical flow tracking on each edge pixel point in the analyzed image in all pelvic floor muscle ultrasound images, determining the optical flow matching points of each edge pixel point, arranging each edge pixel point and its optical flow matching points in time sequence to obtain the tracking sequence of each edge pixel point in the analyzed image; respectively calculating the distances between the first pixel point in the tracking sequence and the remaining pixel points, and taking the maximum value of the distances as the feature screening index of each edge pixel point in the analyzed image; selecting the edge pixel point corresponding to the largest feature screening index on each edge segment in the analyzed image as the feature point; sampling the pixel points in the tracking sequence of each feature point based on the optimized step size of the edge segment where each feature point is located in the analyzed image, and forming the sampling sequence of each feature point by the obtained sampling points; establishing a two-dimensional coordinate system with time as the horizontal axis and the displacement amount of the pixel points as the vertical axis; marking the distances between each pixel point in the sampling sequence and the first pixel point in the two-dimensional coordinate system to obtain the coordinate points of each pixel point; performing curve fitting on the coordinate points in the two-dimensional coordinate system to obtain the displacement curve of each feature point.

[0059] In order to reduce the calculation amount and effectively capture the subtle changes of the pelvic floor muscles, during the movement of the pelvic floor muscles, the edge pixel point with the largest displacement amount on the edge segment is selected as the feature point for optical flow tracking; among them, the feature screening index reflects the maximum displacement amount of the edge pixel point during the movement of the pelvic floor muscles.

[0060] It should be noted that sampling starts from the first pixel point in the tracking sequence of feature points, that is, the first pixel point is the first sampling point, the second sampling point is after the previous sampling point and the interval between them is the optimized step length of the edge segment where the feature point is located in terms of pixels, and so on until the tracking sequence is traversed. The obtained sampling points are arranged in time sequence to obtain the sampling sequence of the feature points. The horizontal axis of the two-dimensional coordinate system represents time, that is, the acquisition moment of the ultrasonic image where the pixel points are located in the tracking sequence of the feature points.

[0061] In an implementation manner of the embodiment of the present invention, the Lucas-Kanade method is selected to perform optical flow tracking on pixel points; the least squares method is selected to perform curve fitting on coordinate points in the two-dimensional coordinate system.

[0062] For analyzing the displacement curve of each feature point in the image, the moment corresponding to the peak point with the largest amplitude and the moment corresponding to the valley point with the smallest amplitude on the displacement curve are successively recorded as the maximum contraction moment and the fully relaxed moment; the amplitude of the maximum contraction moment on the displacement curve is recorded as the maximum contraction distance, the time interval between the maximum contraction moment and the fully relaxed moment is recorded as the diastolic duration, and the ratio of the absolute value of the difference between the amplitudes of the maximum contraction moment and the fully relaxed moment on the displacement curve to the diastolic duration is calculated as the diastolic speed; the maximum contraction distance, diastolic duration, and diastolic speed of the feature point are arranged in order to obtain the feature sequence of the feature point.

[0063] The feature sequences of all feature points in the analyzed image are input into the trained neural network, and the pelvic floor muscle contraction level of the patient is output. In the embodiment of the present invention, the pelvic floor muscle contraction types include: no contraction state, partial contraction state, and full contraction state, which can be set by the implementer according to specific situations.

[0064] In the embodiment of the present invention, the pelvic floor muscle contraction level is labeled through a convolutional neural network. The input of the neural network is the feature sequences of all feature points in the analyzed image of the patient, and the output is the pelvic floor muscle contraction level of the patient.

[0065] Among them, the relevant content of the convolutional neural network includes: the data set of the neural network is divided into a training set and a validation set; the training process of the neural network is the labeling process of the pelvic floor muscle contraction level, and the specific labeling process is: the pelvic floor muscle contraction level of the pelvic floor muscle in the no contraction state is labeled as level 0, the pelvic floor muscle contraction level of the pelvic floor muscle in the partial contraction state is labeled as level 1, and the pelvic floor muscle contraction level of the pelvic floor muscle in the full contraction state is labeled as level 2; the loss function of the neural network is the cross-entropy function. Among them, the convolutional neural network is a well-known technology to those skilled in the art and will not be elaborated here. It should be noted that the higher the pelvic floor muscle contraction level, the better the pelvic floor muscle state of the patient.

[0066] So far, the present invention is completed.

[0067] Embodiment 2: The present invention also provides a schematic diagram of a computer device for an intelligent evaluation device of pelvic floor muscle state using medical imaging analysis. Please refer to Figure 3 , the computer device includes a memory 601, a processor 602, and a computer program 603 stored in the memory 601 and running on the processor 602. When the processor 602 executes the computer program 603, the computer device can execute any one of the intelligent evaluation methods of pelvic floor muscle state using medical imaging analysis introduced above.

[0068] In addition, an embodiment of the present application also protects a device, which may include a memory and a processor. Among them, an executable program code is stored in the memory, and the processor is used to call and execute the executable program code to execute an intelligent evaluation method of pelvic floor muscle state using medical imaging analysis provided by an embodiment of the present application.

[0069] In this embodiment, the device can be divided into functional modules according to the above method examples. For example, it can correspond to each functional module, or two or more functions can be integrated into one processing module. The above integrated module can be implemented in the form of hardware. It should be noted that the division of modules in this embodiment is illustrative, only a logical function division, and there can be other division methods in actual implementation.

[0070] In the case of dividing each module according to each corresponding function, the device may further include a communication module, a signal analysis module, a complexity analysis module, a positioning module, etc. It should be noted that all relevant contents of each step involved in the above method embodiment can be cited in the function description of the corresponding functional module, and will not be repeated here.

[0071] It should be understood that the device provided in this embodiment is used to execute the above intelligent evaluation method of pelvic floor muscle state using medical imaging analysis, so the same effect as the above implementation method can be achieved.

[0072] In the case of adopting an integrated unit, the device may include a processing module and a storage module. Among them, when the device is applied to a device, the processing module can be used to control and manage the actions of the device. The storage module can be used to support the device to execute mutual program codes, etc.

[0073] Among them, the processing module can be a processor or a controller, which can implement or execute various exemplary logical blocks, modules, and circuits included in the disclosure of this application. The processor can also be a combination that implements computing functions, such as including a combination of one or more microprocessors, a combination of digital signal processing (DSP) and a microprocessor, etc. The storage module can be a memory.

[0074] Embodiment 3: This embodiment also provides a computer-readable storage medium, in which computer program code is stored. When the computer program code runs on a computer, the computer is caused to execute the above-mentioned related method steps to implement an intelligent evaluation method for pelvic floor muscle status using medical imaging analysis provided in the above embodiment.

[0075] Embodiment 4: This embodiment also provides a computer program product. When the computer program product runs on a computer, the computer is caused to execute the above-mentioned related steps to implement an intelligent evaluation method for pelvic floor muscle status using medical imaging analysis provided in the above embodiment.

[0076] Among them, the device, computer-readable storage medium, computer program product, or chip provided in this embodiment are all used to execute the corresponding method provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method provided above, and will not be elaborated here.

[0077] In the embodiments provided in this application, it should be understood that the disclosed device and method can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of modules or units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of the device or unit can be in an electrical, mechanical, or other form.

[0078] It should be noted that the above-mentioned sequence of the embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0079] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. An intelligent evaluation method for the state of pelvic floor muscles using medical imaging analysis, characterized in that, The method includes: Obtaining a pelvic floor muscle ultrasound video of a patient; Performing edge detection on each frame of pelvic floor muscle ultrasound image to obtain muscle edges, and dividing the corresponding muscle edges into edge segments according to the clarity of pixel points on each muscle edge; Denoting the first frame of pelvic floor muscle ultrasound image as the analysis image, and obtaining the muscle change complexity of each edge segment in the analysis image according to the differences in bending degree and distribution direction of the corresponding edge segments in every two adjacent frames of pelvic floor muscle ultrasound images, as well as the difference in bending degree of the corresponding edge segment relative to the muscle edge where it is located; Adjusting a preset step size according to the muscle complex change degree of each edge segment in the analysis image and the clarity of the corresponding edge segment in the pelvic floor muscle ultrasound image, and determining the optimized step size of each edge segment in the analysis image; Evaluating the pelvic floor muscle state of the patient according to the position change of sampling points obtained by sampling the pelvic floor muscle ultrasound image using the optimized step size.

2. The intelligent evaluation method for pelvic floor muscle state using medical imaging analysis according to claim 1, characterized in that, The dividing the corresponding muscle edge into edge segments according to the clarity of pixel points on each muscle edge includes: For each edge pixel point in the pelvic floor muscle ultrasound image, dividing the preset neighborhood of the edge pixel point into two local neighborhood regions by using the muscle edge where the edge pixel point is located; obtaining the gray concentration value of the local neighborhood region; Obtaining the point clarity of the edge pixel point according to the difference and gradient value of the gray concentration values of the two local neighborhood regions of the edge pixel point; Clustering the edge pixel points on each muscle edge based on the point clarity to obtain a number of clustering clusters; the edge pixel points on each edge segment are connected and within the same clustering cluster.

3. The intelligent evaluation method for pelvic floor muscle state using medical imaging analysis according to claim 2, wherein, The obtaining the muscle change complexity of each edge segment in the analysis image includes: Obtaining the bending degree of each edge segment; Arranging the corresponding edge segments of each edge segment in the analysis image in all pelvic floor muscle ultrasound images in time sequence to obtain the segment sequence of each edge segment in the analysis image; obtaining the motion deviation degree of each edge segment in the analysis image according to the differences in bending degree and distribution direction of every two adjacent edge segments in the segment sequence; Obtaining the bending degree of the muscle edge where each edge segment is located; selecting the characteristic segment of each edge segment in the analysis image from the segment sequence, and obtaining the muscle change complexity of each edge segment in the analysis image according to the motion deviation degree of each edge segment in the analysis image and the difference in bending degree between the characteristic segment of each edge segment and the muscle edge where the characteristic segment is located.

4. The intelligent evaluation method for pelvic floor muscle state using medical imaging analysis according to claim 3, characterized in that, The obtaining the bending degree of each edge segment includes: Performing linear fitting on the edge pixel points on each edge segment to obtain a distribution line; calculating the average distance of all edge pixel points on each edge segment to the distribution line as the overall spacing; Obtaining the bending degree of the corresponding edge segment according to the average curvature of the edge pixel points on each edge segment and the overall spacing, and both the average curvature and the overall spacing have a positive correlation with the bending degree.

5. The intelligent evaluation method for pelvic floor muscle status using medical imaging analysis according to claim 4, characterized in that, The obtaining the motion deviation degree of each edge segment in the analysis image includes: Obtain the local deviation degree of each edge segment in the segmentation sequence according to the absolute value of the difference between the included angle between each edge segment and its next edge segment of the distribution line in the segmentation sequence of each edge segment in the analysis image and the curvature; the absolute value of the difference between the included angle and the curvature and the local deviation degree are positively correlated; Take the sum of the local deviation degrees of the remaining edge segments except the last edge segment in the segmentation sequence of each edge segment in the analysis image as the motion deviation degree of each edge segment in the analysis image.

6. The intelligent evaluation method for pelvic floor muscle state using medical imaging analysis according to claim 3, wherein The determination of the optimized step size for each edge segment in the analysis image includes: Take the sum of the point sharpness of the edge pixel points on each edge segment as the edge sharpness; Determine the step size adjustment coefficient corresponding to each edge segment according to the muscle change complexity of each edge segment in the analysis image and the edge sharpness of its characteristic segment; the edge sharpness and the step size adjustment coefficient are negatively correlated, and the muscle change complexity and the step size adjustment coefficient are positively correlated; Use the step size adjustment coefficient to weight the preset step size to obtain the optimized step size for each edge segment in the analysis image.

7. An intelligent evaluation method for pelvic floor muscle status using medical imaging analysis according to claim 1, characterized in that, The evaluation of the patient's pelvic floor muscle state according to the position change of the sampling points obtained by sampling the pelvic floor muscle ultrasound image using the optimized step size includes: Perform optical flow tracking on each edge pixel point in the analysis image among all pelvic floor muscle ultrasound images, determine the optical flow matching point of each edge pixel point, and arrange each edge pixel point and its optical flow matching point in time sequence to obtain the tracking sequence of each edge pixel point in the analysis image; Calculate the distances between the first pixel point and the remaining pixel points in the tracking sequence respectively, and take the maximum value of the distances as the feature screening index of each edge pixel point in the analysis image; select the edge pixel point corresponding to the largest feature screening index on each edge segment in the analysis image as the feature point; Based on the optimized step size of the edge segment where each feature point in the analysis image is located, sample the pixel points in the tracking sequence of each feature point, and the obtained sampling points form the sampling sequence of each feature point; Establish a two-dimensional coordinate system with time as the horizontal axis and the displacement amount of the pixel point as the vertical axis; mark the distances between each pixel point in the sampling sequence and the first pixel point in the two-dimensional coordinate system to obtain the coordinate points of each pixel point; perform curve fitting on the coordinate points in the two-dimensional coordinate system to obtain the displacement curve of each feature point; Evaluate the patient's pelvic floor muscle state based on the displacement curve of the feature points in the analysis image.

8. The intelligent evaluation method for pelvic floor muscle state using medical imaging analysis according to claim 3, characterized in that, The characteristic segment is the edge segment corresponding to the maximum curvature in the segmentation sequence of each edge segment in the analysis image.

9. The intelligent evaluation method for pelvic floor muscle state using medical imaging analysis according to claim 1, characterized in that, The muscle edge is the edge obtained by curve fitting the edge pixel points detected by edge detection for each frame of pelvic floor muscle ultrasound image.

10. The intelligent evaluation method for pelvic floor muscle status using medical imaging analysis according to claim 2, characterized in that, The gray concentration value is equal to the gray mean value of all pixel points in the local neighborhood area.

Images