A fracture identification method for tension detection of medical sutures

Abstract

The present invention relates to the field of image processing technology, and in particular to a fracture identification method suitable for tension detection of medical sutures. The method comprises: obtaining a plurality of monitoring images of medical sutures under different tensions; matching the pixel points in the monitoring image except the initial edge line with the pixel points on the initial edge line to obtain the virtual edge line corresponding to each initial edge line; according to the relative position distribution and quantity distribution between the initial edge line and the corresponding virtual edge line in each monitoring image under each tension, obtaining the blur degree of each monitoring image under each tension and screening the target image under each tension; based on the size distribution of the blur degree of the monitoring image under each tension, determining the degree of jitter under each tension; comprehensively determining the degree of jitter and the distribution of the broken braided thin wires, determining the overall quality evaluation value, and then screening out the fracture segment. The present invention improves the fracture identification accuracy of medical sutures.

Description

Technical Field

[0001] The invention relates to the technical field of image processing, and in particular to a fracture identification method suitable for tension detection of medical sutures. Background Art

[0002] Suture tension refers to the pulling force that the suture bears when closing a wound. Appropriate tension can help align the wound edges, reduce the risk of wound dehiscence, and help reduce excessive scarring, making the healed skin smoother. The suture tension tester can be used to evaluate the tensile strength of sutures of different materials to ensure that the sutures produced meet standard requirements.

[0003] The traditional method determines the tension of the suture by its deformation under tension, but does not take into account the influence of the occlusion of the broken thread and the shaking of the suture after tension on the deformation of the suture, resulting in low accuracy in the fracture identification of medical sutures. Summary of the invention

[0004] In order to solve the problem of low accuracy in the existing methods for identifying fractures of medical sutures, the purpose of the present invention is to provide a fracture identification method suitable for tension detection of medical sutures, and the technical scheme adopted is as follows:

[0005] The present invention provides a fracture identification method suitable for tension detection of medical sutures, the method comprising the following steps:

[0006] Acquire several monitoring images of medical sutures under different tensions;

[0007] Perform edge detection on each monitoring image to obtain an initial edge line; match the pixel points in the monitoring image except the initial edge line with the pixel points on the initial edge line to obtain the virtual edge line corresponding to each initial edge line; obtain the blur degree of each monitoring image under each tension according to the relative position distribution and quantity distribution between the initial edge line and the corresponding virtual edge line in each monitoring image under each tension, and screen the target image under each tension;

[0008] Based on the distribution of the blur degree of the monitoring image under each tension, the jitter degree of the medical suture under each tension is determined; based on the similarity between different initial edge lines in the target image under each tension, the broken braided thin wires are screened; the maximum tensile force that the medical suture can withstand is determined based on the comprehensive distribution of the jitter degree and the number of broken braided thin wires; based on the distribution of the broken braided thin wires in the target image under the maximum tensile force, the overall quality evaluation value of each section of the medical suture is determined;

[0009] The overall quality evaluation value is used to screen broken segments on medical sutures.

[0010] Preferably, the blur degree of each monitoring image under each pulling force is obtained according to the relative position distribution and quantity distribution between the initial edge line and the corresponding virtual edge line in each monitoring image under each pulling force, including:

[0011] Obtaining a blur degree of the candidate image according to the number of virtual edge lines corresponding to each initial edge line in the candidate image and the maximum value of the distances between all initial edge lines in the candidate image and their corresponding virtual edge lines, wherein the number of virtual edge lines and the maximum value of the distances are both positively correlated with the blur degree;

[0012] The candidate image is any monitoring image under any tension.

[0013] Preferably, screening the target image under each tension includes:

[0014] For any pulling force, the monitoring image with the smallest blur degree among all monitoring images under the pulling force is determined as the target image under the pulling force.

[0015] Preferably, the determining the shaking degree of the medical suture thread under each pulling force based on the size distribution of the blur degree of the monitoring image under each pulling force includes:

[0016] Counting respectively a first number of monitoring images having a blur degree less than a preset blur degree among all monitoring images under each pulling force;

[0017] For any pulling force: the degree of jitter of the medical suture thread under this pulling force is obtained based on the maximum value of the blur degree in all monitoring images under this pulling force, the first quantity corresponding to this pulling force and the maximum value of the first quantity corresponding to all pulling forces.

[0018] Preferably, screening the broken braided thin wires according to the similarity between different initial edge lines in the target image under each pulling force comprises:

[0019] For any tension:

[0020] Based on the similarity between every two initial edge lines in the target image under the tension, all initial edge lines in the target image under the tension are matched to obtain a plurality of groups;

[0021] Calculate the average distance between each initial edge line and the matching pixels on other matching initial edge lines in its group;

[0022] According to the difference between the distance between each initial edge line and the pixel points on other initial edge lines in the group to which it belongs and the average distance, the possibility of each braided thin line being broken is calculated; wherein each braided thin line is an initial edge line;

[0023] If the possibility degree is greater than a preset possibility degree threshold, it is determined that the corresponding braided thin wire is a broken braided thin wire.

[0024] Preferably, the maximum tensile force of the medical suture is determined by combining the degree of vibration and the number of broken braided threads, including:

[0025] For any tension:

[0026] According to the number of broken braided thin threads in the target image under the tension, the wire diameter width of the medical suture under the tension, and the degree of jitter of the medical suture under the tension, the insufficient bearing degree of the medical suture under the tension is obtained, wherein the number of broken braided thin threads and the degree of jitter are both positively correlated with the insufficient bearing degree, and the wire diameter width is negatively correlated with the insufficient bearing degree;

[0027] The first difference of the insufficient bearing degree of the medical suture thread under every two adjacent pulling forces is calculated; and the smallest pulling force of the two pulling forces corresponding to the maximum value of the first difference is determined as the maximum bearing pulling force.

[0028] Preferably, the overall quality evaluation value of each section of medical suture thread is determined according to the distribution of the broken braided threads in the target image under the maximum tensile force, including:

[0029] Divide the medical suture line in the target image under the maximum tensile force into multiple segments;

[0030] For any section of medical suture thread, the first distance between each position of the broken braided thread in the section of medical suture thread in the target image under each tension and the broken braided thread closest to it is calculated; based on the first distance, the maximum tolerable tension and the number of broken braided threads in the target image under the maximum tolerable tension, the overall quality evaluation value of the section of medical suture thread is calculated.

[0031] Preferably, the calculating of the overall quality evaluation value of the medical suture according to the first distance, the maximum tensile force and the number of braided thin threads broken in the target image under the maximum tensile force comprises:

[0032] For any length of medical suture:

[0033] Calculating respectively a first average value of first distances between all positions of the broken braided thin threads in the section of the medical suture thread and the broken braided thin threads closest to the positions in the target image under each pulling force;

[0034] According to the difference between the first distance corresponding to each position of the broken braided thread in the section of medical suture thread in the target image under each tension, and the first average value corresponding to the broken braided thread, the maximum tension and the number of broken braided threads in the target image under the maximum tension, the overall quality evaluation value of the section of medical suture thread is obtained, the maximum tension is positively correlated with the overall quality evaluation value, and the difference between the first distance and the first average value corresponding to the broken braided thread and the number of broken braided threads in the target image under the maximum tension are negatively correlated with the overall quality evaluation value.

[0035] Preferably, the method of screening broken segments on a medical suture using the overall quality evaluation value comprises:

[0036] A medical suture line segment whose overall quality evaluation value is less than a preset evaluation threshold is regarded as a broken segment on the medical suture line.

[0037] Preferably, matching the pixel points in the monitoring image other than the initial edge line with the pixel points on the initial edge line to obtain the virtual shadow edge line corresponding to each initial edge line includes:

[0038] For the candidate image: a template matching algorithm is used to match the pixel points in the candidate image except the initial edge line with the pixel points on the initial edge line, and the line formed by each set of successfully matched pixel points is used as each virtual edge line corresponding to each initial edge line.

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

[0040] The present invention firstly utilizes a plurality of different pulling forces to pull the medical sutures, and collects a plurality of monitoring images of the medical sutures under each pulling force. Considering that a certain pulling force is applied to the medical sutures during the image acquisition process, the existence of the pulling force will cause the medical sutures to shake, which will further cause the collected monitoring images of the medical sutures to be blurred, thus affecting the subsequent fracture recognition results. The present invention calculates the blur degree of each monitoring image according to the relative position distribution and quantity distribution between the initial edge line and the corresponding virtual edge line in each monitoring image under each pulling force, and screens The target image under each pulling force was obtained. The target image was less affected by jitter and was suitable for fracture identification. Then, based on the size distribution of the blur degree of the monitoring image under each pulling force, the jitter degree of the medical suture thread under each pulling force was evaluated, and the broken braided threads were screened according to the similarity between different initial edge lines in the target image under each pulling force. The overall quality of each section of medical suture thread was evaluated based on the comprehensive jitter degree and the distribution of the broken braided threads, and then the broken sections on the medical suture thread were screened out, thereby improving the accuracy of medical suture thread fracture identification. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0042] Figure 1 A flow chart of a fracture identification method applicable to tension detection of medical sutures provided by an embodiment of the present invention;

[0043] Figure 2 A structural block diagram of a fracture identification system suitable for tension detection of medical sutures provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0044] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, a fracture identification method suitable for tension detection of medical sutures proposed in accordance with the present invention is described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0045] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0046] A specific scheme of a fracture identification method suitable for tension detection of medical sutures provided by the present invention is described in detail below with reference to the accompanying drawings.

[0047] Embodiment of the fracture identification method applicable to the tension detection of medical sutures:

[0048] The specific scenario targeted by this embodiment is: when identifying the fracture of medical sutures, the medical sutures are pulled with different pulling forces, and monitoring images of the medical sutures under different pulling forces are collected, and the obtained monitoring images are screened by fuzzy analysis, and the degree of jitter of the medical sutures is judged. The broken braided thin wires are judged by the uniformity of the spacing between the braided thin wires in the clear image, and the overall quality of each section of the medical suture is evaluated, and then the fracture position on the medical suture is screened.

[0049] This embodiment proposes a fracture identification method suitable for tension detection of medical sutures, such as Figure 1 As shown, the fracture identification method applicable to the tension detection of medical sutures of this embodiment comprises the following steps:

[0050] Step S1, obtaining a plurality of monitoring images of the medical suture under different tensions.

[0051] In this embodiment, a medical suture with a length of 20 cm is first cut, and then fixed between two clamps of the tester. Starting with a tension of 0, the tension of the suture is increased every 1 minute, by 0.1N each time, until the suture is completely broken. In this process, a preset number of monitoring images of the medical suture are taken with a high-definition camera after each change in tension. In this embodiment, the preset number is 300. In specific applications, the implementer can set it according to specific circumstances. It should be noted that the diameter of the medical suture in this embodiment is 0.1 mm.

[0052] So far, multiple monitoring images of the medical suture under each tension have been obtained.

[0053] Step S2, performing edge detection on each monitoring image to obtain an initial edge line; matching the pixel points in the monitoring image except the initial edge line with the pixel points on the initial edge line to obtain a virtual edge line corresponding to each initial edge line; according to the relative position distribution and quantity distribution between the initial edge line and the corresponding virtual edge line in each monitoring image under each pulling force, obtaining the blur degree of each monitoring image under each pulling force and screening the target image under each pulling force.

[0054] As the tension increases, the greater the tension on the medical suture, the more the suture will shake, and the degree of shaking will gradually increase. In order to avoid the suture shaking causing the collected monitoring images to be blurred when the tension is large, and to affect the subsequent identification of the suture breakage based on the deformation of the medical suture under different tensions, all monitoring images should be screened according to the blurriness of the obtained images to obtain high-definition images.

[0055] The Sobel edge detection algorithm is used to perform edge detection on each monitoring image, and the edge line obtained at this time is recorded as the initial edge line, that is, multiple initial edge lines in each monitoring image are obtained, and the initial edge line can be detected by the edge detection algorithm, so the initial edge line belongs to the clear edge line in the image. The Sobel edge detection algorithm is a prior art and will not be described in detail here.

[0056] Next, this embodiment is described by taking any surveillance image as an example, and other surveillance images can be processed by the method provided in this embodiment.

[0057] Specifically, any monitoring image under any pulling force is recorded as a candidate image, and the pixel points in the candidate image except the initial edge line are matched with the pixel points on the initial edge line by using a template matching algorithm, and the line formed by each set of successfully matched pixel points is used as the ghost edge line corresponding to each initial edge line. It should be noted that: an initial edge line may have multiple corresponding ghost edge lines. According to the number of ghost edge lines corresponding to each initial edge line in the candidate image, and the maximum value of the distance between all the initial edge lines in the candidate image and the corresponding ghost edge lines, the blur degree of the candidate image is obtained, and the number of the ghost edge lines and the maximum value of the distance are both positively correlated with the blur degree.

[0058] Among them, the positive correlation means that the dependent variable will increase as the independent variable increases, and the dependent variable will decrease as the independent variable decreases. It can be an additive relationship, a multiplicative relationship, etc., which is determined by practical applications.

[0059] In this embodiment, a specific calculation formula for the blur degree is given, and the calculation formula for the blur degree of the j-th monitoring image under the i-th pulling force is:

[0060]

[0061] in, represents the blur degree of the jth monitoring image under the i-th tension, represents the number of initial edge lines in the jth monitoring image under the i-th tension, represents the number of virtual edge lines corresponding to the g-th initial edge line in the j-th monitoring image under the i-th tension, represents the maximum value of the distance between all initial edge lines and their corresponding virtual edge lines in the jth monitoring image under the i-th tension, represents the maximum value of the distance between all initial edge lines and their corresponding virtual edge lines in all monitoring images under all tensions, and sigmoid() represents the normalization function, which is used to convert The value of falls into (0, 1).

[0062] Used for After normalization, the larger the ratio is, the more blurred the jth monitoring image under the i-th tension is; Characterizes the average value of the number of ghost edge lines corresponding to all initial edge lines in the jth monitoring image under the i-th tension. When the maximum value of the distance between all initial edge lines and their corresponding ghost edge lines in the jth monitoring image under the i-th tension is larger and the average value of the number of ghost edge lines corresponding to all initial edge lines in the jth monitoring image under the i-th tension is also larger, it means that the jth monitoring image under the i-th tension is more blurred, that is, the blur degree of the jth monitoring image under the i-th tension is greater.

[0063] By using the above method, the blur degree of each monitoring image under each tension can be obtained. The smaller the blur degree, the clearer the corresponding monitoring image is, and the more suitable it is for subsequent analysis. Therefore, for any tension, the monitoring image with the smallest blur degree among all monitoring images under the tension is determined as the target image under the tension. By using this method, the target image under each tension can be screened out. The target image is the monitoring image with the highest clarity among the captured images, which can be used for subsequent identification of fracture areas.

[0064] Step S3, based on the distribution of the degree of blur of the monitoring image under each tension, determine the degree of jitter of the medical suture under each tension; screen the broken braided threads according to the similarity between different initial edge lines in the target image under each tension; determine the maximum tensile force that the medical suture can withstand based on the comprehensive distribution of the degree of jitter and the number of broken braided threads; determine the overall quality evaluation value of each section of medical suture according to the distribution of the broken braided threads in the target image under the maximum tensile force.

[0065] When the number of virtual images in a single high-definition image obtained is greater and the distance between the farthest virtual image and its corresponding original boundary is farther, the blurriness of the obtained high-definition image is greater and the jitter degree of the medical suture line is greater.

[0066] The first number of monitoring images with blur levels less than a preset blur level in all monitoring images under each pulling force is counted respectively; in this embodiment, the preset blur level threshold is 0.5, and in specific applications, the implementer can set it according to specific circumstances.

[0067] Next, this embodiment is described by taking a tension as an example, and other tensions can be processed by the method provided in this embodiment.

[0068] Specifically, for any pulling force: the degree of jitter of the medical suture thread under this pulling force is obtained based on the maximum value of the blur degree in all monitoring images under this pulling force, the first quantity corresponding to this pulling force, and the maximum value of the first quantity corresponding to all pulling forces.

[0069] In this embodiment, a specific calculation formula for the shaking degree is given, and the shaking degree of the medical suture thread under the i-th tension can be expressed as:

[0070]

[0071] in, It indicates the shaking degree of the medical suture under the i-th tension. represents the maximum value of blur degree in all monitoring images under the i-th tension, Indicates the maximum value of the number of monitoring images whose blurring degree is less than the preset blurring degree among all monitoring images under all pulling forces, that is, the maximum value of the first number corresponding to all pulling forces; It represents the number of monitoring images whose blurring degree is less than the preset blurring degree among all monitoring images under the i-th pulling force, that is, the first number corresponding to the i-th pulling force.

[0072] In this embodiment, the preset blur level is 0.5. In specific applications, the implementer can set it according to specific circumstances. The larger the value of , the fewer images with smaller blurring degrees among all target images of the medical suture line under the i-th tension, reflecting that the degree of jitter of the medical suture line is greater.

[0073] The medical suture thread targeted by this embodiment is a braided thread made of synthetic fiber (such as polyester, polyolefin, polytetrafluoroethylene, polyurethane, etc.). When a large pulling force is applied, the braided thread may break, and the broken braided thread may block the texture distribution of the unbroken braided thread, and the broken braided thread protrudes relative to the unbroken thread, which will affect the measurement of the diameter of the medical suture thread.

[0074] The braided medical suture itself has a certain distribution law when it is braided. When adjacent braided thin wires are not broken, the spacing at different positions is consistent. When a large tensile force is applied, the braided thin wires that bear the larger tensile force will break first, and after breaking, they will no longer meet the distribution law when they are not broken. Therefore, whether the braided thin wires are broken is judged based on the similarity between different initial edge lines.

[0075] Next, this embodiment is still described by taking a tension as an example, and other tensions can be processed by the method provided in this embodiment.

[0076] For any pulling force: based on the similarity between each two initial edge lines in the target image under this pulling force, a template matching algorithm is used to match all the initial edge lines in the target image under this pulling force to obtain multiple groups; the template matching algorithm is a prior art and will not be described in detail here. The average distance between each initial edge line and the matching pixel points on other matching initial edge lines in its group is calculated; based on the difference between the distance between each initial edge line and the pixel points on other initial edge lines in its group and the average distance, the possibility of each braided thin wire being broken is calculated; wherein each braided thin wire is an initial edge line.

[0077] In this embodiment, a specific calculation formula for the possibility degree is given. The possibility degree of the kth braided thin wire being broken in the target image under this tension can be expressed as:

[0078]

[0079] in, It indicates the possibility of breaking the kth braided wire in the target image under this tension. represents the number of initial edge lines in the group where the k-th braided thin wire belongs, U represents the number of pixel points on the k-th braided thin wire in the target image under this tension, represents the distance between the u-th pixel point on the k-th braided wire in the target image under this tension and the matching pixel point on the s-th initial edge line excluding the k-th braided wire in the group where the k-th braided wire belongs, represents the average distance between the pixel points on the k-th braided wire and the matching pixel points on the s-th initial edge line in its group except the k-th braided wire. represents the absolute value sign, and sigmoid() represents the normalization function.

[0080] To reflect and The greater the absolute value, the greater the difference between the two. The fewer the initial edge lines in the group where the k-th braided thin wire belongs, the less the distribution of the k-th braided thin wire does not conform to the braiding law, and the more likely the k-th braided thin wire has been broken. The larger it is, the more likely it is that the kth braided wire in the target image will be broken under this tension.

[0081] By using the above method, the possibility of each braided thin wire being broken in the target image under the tension can be obtained. If the possibility is greater than the preset possibility threshold, the corresponding braided thin wire is determined to be a broken braided thin wire. In this embodiment, the preset possibility threshold is 0.7. In specific applications, the implementer can set it according to specific circumstances.

[0082] Next, based on the number of broken braided threads in the target image under this type of tension, the wire width of the medical suture thread under this type of tension, and the degree of jitter of the medical suture thread under this type of tension, the degree of insufficient tolerance of the medical suture thread under each type of tension is obtained. The number of broken braided threads and the degree of jitter are positively correlated with the degree of insufficient tolerance, and the wire width is negatively correlated with the degree of insufficient tolerance.

[0083] Among them, a positive correlation means that the dependent variable will increase as the independent variable increases, and the dependent variable will decrease as the independent variable decreases. It can be an additive relationship, a multiplicative relationship, etc., which is determined by actual application; a negative correlation means that the dependent variable will decrease as the independent variable increases, and the dependent variable will increase as the independent variable decreases. It can be a subtractive relationship, a division relationship, etc., which is determined by actual application.

[0084] In this embodiment, a specific calculation formula for the insufficient bearing degree is given, and the insufficient bearing degree of the medical suture thread under the i-th tension can be expressed as:

[0085]

[0086] in, It indicates the insufficient bearing degree of medical suture under the i-th tension. It indicates the shaking degree of the medical suture under the i-th tension. It indicates the maximum value of the vibration of medical suture under all tensions. represents the number of broken braided threads in the target image under the i-th tension, Represents the wire diameter width of the medical suture under the i-th tension.

[0087] It should be noted that the method for obtaining the wire diameter width of the medical suture thread under the i-th tension is: obtain the top edge line and the bottom edge line of the medical suture thread in the target image under the i-th tension, and the pixel points in these two edge lines correspond one to one, that is, for each pixel point in the top edge line of the medical suture thread, there is a point in the bottom edge line corresponding to it, that is, multiple pairs of pixel points are formed, and the distance between the two pixel points in each pair of pixel points is calculated respectively, and the average value of all distances is used as the wire diameter width of the medical suture thread under the i-th tension.

[0088] When the wire diameter width of the medical suture thread under the i-th tension is narrower, more braided thin wires are broken in the target image under the i-th tension, and the degree of jitter of the medical suture thread under the i-th tension is greater, it means that the deformation degree of the medical suture thread is greater, and the force should not be further applied, that is, the degree of insufficient bearing capacity of the medical suture thread under the i-th tension is greater.

[0089] By adopting the above method, the degree of insufficient bearing of the medical suture under each tension can be obtained. According to the order of tension from small to large, the difference between the degree of insufficient bearing of the medical suture under each two adjacent tensions is calculated respectively, and recorded as the first difference. It should be noted that when calculating the difference between the degree of insufficient bearing of the medical suture under each two adjacent tensions, the difference is obtained by subtracting the degree of insufficient bearing of the medical suture under the previous tension from the degree of insufficient bearing of the medical suture under the latter tension of the two adjacent tensions. Each two adjacent tensions correspond to a first difference. The smallest tension of the two tensions corresponding to the maximum value of the first difference is determined as the maximum bearing tension of the medical suture.

[0090] The medical suture line in the target image under the maximum tensile force is divided into multiple segments; the specific number of divisions is set according to the specific situation.

[0091] For any section of medical suture, the distance between each position of the broken braided thin wire in the section of medical suture in the target image under each tension and the nearest broken braided thin wire is calculated, and the distance is recorded as the first distance. The first average value of the first distance between all positions of the broken braided thin wire in the section of medical suture and the nearest broken braided thin wire in the target image under each tension is calculated respectively. According to the difference between the first distance corresponding to each position of the broken braided thin wire in the section of medical suture in the target image under each tension and the first average value corresponding to the broken braided thin wire, the maximum tension and the number of broken braided thin wires in the target image under the maximum tension, the overall quality evaluation value of the section of medical suture is obtained, the maximum tension is positively correlated with the overall quality evaluation value, the difference between the first distance and the first average value corresponding to the broken braided thin wire and the number of broken braided thin wires in the target image under the maximum tension are both negatively correlated with the overall quality evaluation value.

[0092] In this embodiment, a specific calculation formula for the overall quality evaluation value is given, and the overall quality evaluation value of the p-th section of medical suture thread can be expressed as:

[0093]

[0094] in, represents the overall quality evaluation value of the p-th segment of medical suture, W represents the maximum tensile force, I represents the number of types of pressure, and T represents the number of locations on the braided thin wire that is broken in the p-th segment of medical suture in the target image under the i-th tensile force; represents the distance between the t-th position of the broken braided thin thread in the p-th section of the medical suture in the target image under the i-th tension and the closest broken braided thin thread, that is, the first distance corresponding to the t-th position of the broken braided thin thread in the p-th section of the medical suture in the target image under the i-th tension; represents the average value of the distances between all positions of the broken braided thin wires in the p-th medical suture line in the target image under the i-th tension and the closest broken braided thin wire, that is, the average value of the first distances corresponding to all positions of the broken braided thin wires in the p-th medical suture line in the target image under the i-th tension; Indicates the number of broken braided threads in the target image under the maximum tensile force; Indicates the preset first adjustment parameter.

[0095] In this embodiment, the preset first adjustment parameter is introduced into the calculation formula of the overall quality evaluation value to prevent the denominator from being 0. In this embodiment, the preset first adjustment parameter is 0.01. In specific applications, the implementer can set it according to specific circumstances.

[0096] express and The smaller the absolute value, the smaller the difference between the two. The smaller the value is, the more uniform the breaking positions of the thin lines in the suture when the i-th tension is applied to the medical suture, the better the quality of the suture, and the greater the overall quality evaluation value.

[0097] By using the above method, the overall quality evaluation value of each section of medical suture can be obtained.

[0098] Step S4, using the overall quality evaluation value to screen broken segments on the medical suture.

[0099] In step S3, the present embodiment calculates the overall quality evaluation value of each segment of medical suture. The larger the overall quality evaluation value, the better the quality of the corresponding medical suture segment. Therefore, the medical suture segment with an overall quality evaluation value less than a preset evaluation threshold is regarded as a broken segment on the medical suture. In the present embodiment, the preset evaluation threshold is 0.75. In specific applications, the implementer can set it according to specific circumstances.

[0100] By adopting the above method, all the broken segments on the medical suture can be screened out, and all the screened broken segments on the medical suture can be marked out for display.

[0101] This embodiment first uses multiple different pulling forces to pull the medical sutures, and collects multiple monitoring images of the medical sutures under each pulling force. Considering that a certain pulling force is applied to the medical sutures during the image acquisition process, the presence of the pulling force will cause the medical sutures to shake, which will make the collected monitoring images of the medical sutures blurred, affecting the subsequent fracture recognition results. This embodiment calculates the blur degree of each monitoring image based on the relative position distribution and quantity distribution between the initial edge line and the corresponding virtual edge line in each monitoring image under each pulling force, and screens the image. A target image under each pulling force was selected. The target image was less affected by jitter and was suitable for fracture identification. Then, based on the size distribution of the blur degree of the monitoring image under each pulling force, the jitter degree of the medical suture under each pulling force was evaluated, and the broken braided threads were screened according to the similarity between different initial edge lines in the target image under each pulling force. The overall quality of each section of medical suture was evaluated based on the comprehensive jitter degree and the distribution of the broken braided threads, and then the broken sections on the medical suture were screened out, thereby improving the accuracy of medical suture fracture identification.

[0102] Embodiment of a fracture identification system suitable for tension detection of medical sutures:

[0103] like Figure 2 As shown, the figure shows a structural block diagram of a fracture identification system suitable for medical suture tension detection, the system includes an image acquisition module, a target image screening module, an evaluation module and a fracture segment screening module.

[0104] Among them, the image acquisition module is used to obtain several monitoring images of the medical suture under different tensions;

[0105] The target image screening module is used to perform edge detection on each monitoring image to obtain the initial edge line; match the pixel points in the monitoring image except the initial edge line with the pixel points on the initial edge line to obtain the virtual edge line corresponding to each initial edge line; according to the relative position distribution and quantity distribution between the initial edge line and the corresponding virtual edge line in each monitoring image under each tension, obtain the blur degree of each monitoring image under each tension and screen the target image under each tension;

[0106] An evaluation module is used to determine the degree of jitter of the medical suture under each tension based on the distribution of the degree of blur of the monitoring image under each tension; screen the broken braided thin wires according to the similarity between different initial edge lines in the target image under each tension; determine the maximum tolerable tension of the medical suture based on the comprehensive jitter degree and the number distribution of the broken braided thin wires; determine the overall quality evaluation value of each section of the medical suture according to the distribution of the broken braided thin wires in the target image under the maximum tolerable tension;

[0107] The broken segment screening module is used to screen the broken segments on the medical suture thread by using the overall quality evaluation value.

[0108] It should be understood that Figure 2 The structural block diagram of a fracture identification system suitable for tension detection of medical sutures and its modules shown can be implemented in various ways. For example, in some embodiments, the system and its modules can be implemented by hardware, software, or a combination of software and hardware. Among them, the hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated design hardware. Those skilled in the art can understand that the above-mentioned methods and systems can be implemented using computer executable instructions and / or included in processor control codes, such as carrier media such as disks, CDs or DVD-ROMs, programmable memories such as read-only memories (firmware), or data carriers such as optical or electronic signal carriers. Such codes are provided on. The system and its modules of this specification can not only be implemented by hardware circuits such as ultra-large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., but can also be implemented by software executed by various types of processors, and can also be implemented by a combination of the above-mentioned hardware circuits and software (for example, firmware).

[0109] For more details about the above modules, please refer to other places in this manual and will not be repeated here.

[0110] In other embodiments, a fracture identification device suitable for tension detection of medical sutures is also provided, comprising a memory and a processor. The memory is used to store executable program code, and the processor is used to call and run the executable program code from the memory, so that the device executes the above-mentioned fracture identification method suitable for tension detection of medical sutures. The device can be specifically a chip, a component or a module, and the chip may include a connected processor and a memory; wherein the memory is used to store instructions, and when the processor calls and executes the instructions, the chip can execute the fracture identification method suitable for tension detection of medical sutures provided in the above-mentioned embodiment.

[0111] In other embodiments, a computer program product is also provided. When the computer program product is run on a computer, the computer executes the above-mentioned related steps to implement the fracture identification method suitable for medical suture tension detection provided in the above-mentioned embodiment.

[0112] In other embodiments, a computer-readable storage medium is also provided, in which a computer program code is stored. When the computer program code is executed on a computer, the computer executes the above-mentioned related method steps to implement the fracture identification method suitable for medical suture tension detection provided in the above-mentioned embodiment.

[0113] Among them, the provided system, electronic device, computer program product, and computer-readable storage medium are all used to execute the corresponding methods provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding methods provided above, and will not be repeated here.

[0114] It should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A fracture identification method suitable for tension detection of medical sutures, characterized in that: The method comprises the following steps: Acquire several monitoring images of medical sutures under different tensions; Perform edge detection on each monitoring image to obtain an initial edge line; match the pixel points in the monitoring image except the initial edge line with the pixel points on the initial edge line to obtain the virtual edge line corresponding to each initial edge line; obtain the blur degree of each monitoring image under each tension according to the relative position distribution and quantity distribution between the initial edge line and the corresponding virtual edge line in each monitoring image under each tension, and screen the target image under each tension; Based on the distribution of the blur degree of the monitoring image under each tension, the jitter degree of the medical suture under each tension is determined; based on the similarity between different initial edge lines in the target image under each tension, the broken braided thin wires are screened; the maximum tensile force that the medical suture can withstand is determined based on the comprehensive distribution of the jitter degree and the number of broken braided thin wires; based on the distribution of the broken braided thin wires in the target image under the maximum tensile force, the overall quality evaluation value of each section of the medical suture is determined; The overall quality evaluation value is used to screen broken segments on medical sutures.

2. The fracture identification method for medical suture tension detection according to claim 1, characterized in that: According to the relative position distribution and quantity distribution between the initial edge line and the corresponding virtual edge line in each monitoring image under each pulling force, the blur degree of each monitoring image under each pulling force is obtained, including: Obtaining a blur degree of the candidate image according to the number of virtual edge lines corresponding to each initial edge line in the candidate image and the maximum value of the distances between all initial edge lines in the candidate image and their corresponding virtual edge lines, wherein the number of virtual edge lines and the maximum value of the distances are both positively correlated with the blur degree; The candidate image is any monitoring image under any tension.

3. The fracture identification method for medical suture tension detection according to claim 1, characterized in that: Filter target images at each tension, including: For any pulling force, the monitoring image with the smallest blur degree among all monitoring images under the pulling force is determined as the target image under the pulling force.

4. The fracture identification method for medical suture tension detection according to claim 1, characterized in that: The determining of the shaking degree of the medical suture thread under each pulling force based on the size distribution of the blur degree of the monitoring image under each pulling force includes: Counting respectively a first number of monitoring images having a blur degree less than a preset blur degree among all monitoring images under each pulling force; For any pulling force: the degree of jitter of the medical suture thread under this pulling force is obtained based on the maximum value of the blur degree in all monitoring images under this pulling force, the first quantity corresponding to this pulling force and the maximum value of the first quantity corresponding to all pulling forces.

5. The fracture identification method for medical suture tension detection according to claim 1, characterized in that: The method of screening the broken braided thin wires according to the similarity between different initial edge lines in the target image under each pulling force includes: For any tension: Based on the similarity between every two initial edge lines in the target image under the tension, all initial edge lines in the target image under the tension are matched to obtain a plurality of groups; Calculate the average distance between each initial edge line and the matching pixels on other matching initial edge lines in its group; According to the difference between the distance between each initial edge line and the pixel points on other initial edge lines in the group to which it belongs and the average distance, the possibility of each braided thin line being broken is calculated; wherein each braided thin line is an initial edge line; If the possibility degree is greater than a preset possibility degree threshold, it is determined that the corresponding braided thin wire is a broken braided thin wire.

6. The fracture identification method for medical suture tension detection according to claim 5, characterized in that: The comprehensive shaking degree and the number distribution of the broken braided thin threads determine the maximum tensile force of the medical suture thread, including: For any tension: According to the number of broken braided thin threads in the target image under the tension, the wire diameter width of the medical suture under the tension, and the degree of jitter of the medical suture under the tension, the insufficient bearing degree of the medical suture under the tension is obtained, wherein the number of broken braided thin threads and the degree of jitter are both positively correlated with the insufficient bearing degree, and the wire diameter width is negatively correlated with the insufficient bearing degree; The first difference of the insufficient bearing degree of the medical suture thread under every two adjacent pulling forces is calculated; and the smallest pulling force of the two pulling forces corresponding to the maximum value of the first difference is determined as the maximum bearing pulling force.

7. The fracture identification method for medical suture tension detection according to claim 1, characterized in that: The overall quality evaluation value of each section of medical suture thread is determined according to the distribution of the broken braided thin threads in the target image under the maximum tensile force, including: Divide the medical suture line in the target image under the maximum tensile force into multiple segments; For any section of medical suture thread, the first distance between each position of the broken braided thread in the section of medical suture thread in the target image under each tension and the broken braided thread closest to it is calculated; based on the first distance, the maximum tolerable tension and the number of broken braided threads in the target image under the maximum tolerable tension, the overall quality evaluation value of the section of medical suture thread is calculated.

8. The fracture identification method for medical suture tension detection according to claim 7, characterized in that: The step of calculating the overall quality evaluation value of the medical suture according to the first distance, the maximum tensile force, and the number of broken braided threads in the target image under the maximum tensile force comprises: For any length of medical suture: Calculating respectively a first average value of first distances between all positions of the broken braided thin threads in the section of the medical suture thread and the broken braided thin threads closest to the positions in the target image under each pulling force; According to the difference between the first distance corresponding to each position of the broken braided thread in the section of medical suture thread in the target image under each tension, and the first average value corresponding to the broken braided thread, the maximum tension and the number of broken braided threads in the target image under the maximum tension, the overall quality evaluation value of the section of medical suture thread is obtained, the maximum tension is positively correlated with the overall quality evaluation value, and the difference between the first distance and the first average value corresponding to the broken braided thread and the number of broken braided threads in the target image under the maximum tension are negatively correlated with the overall quality evaluation value.

9. The fracture identification method for medical suture tension detection according to claim 1, characterized in that: The method of screening the broken segments on the medical suture thread by using the overall quality evaluation value comprises: A medical suture line segment whose overall quality evaluation value is less than a preset evaluation threshold is regarded as a broken segment on the medical suture line.

10. The fracture identification method for medical suture tension detection according to claim 2, characterized in that: The step of matching the pixel points in the monitoring image except the initial edge line with the pixel points on the initial edge line to obtain the virtual shadow edge line corresponding to each initial edge line includes: For the candidate image: a template matching algorithm is used to match the pixel points in the candidate image except the initial edge line with the pixel points on the initial edge line, and the line formed by each set of successfully matched pixel points is used as each virtual edge line corresponding to each initial edge line.

Images