An ultrasonic assessment method for wound trauma of free flap transplantation

Abstract

The present application relates to the technical field of medical information processing, in particular to a kind of free flap grafting wound injury ultrasound evaluation method, comprising: the ultrasound gray scale image obtained is carried out block processing to obtain each enhanced analysis candidate block, block processing helps to improve the contrast of fat area and edema area;By fat feature screening parameter screening obtains several enhanced analysis target blocks, to a certain extent reduce image data analysis amount, improve the efficiency of wound injury ultrasound evaluation;Blood vessel ultrasound frequency shift characteristic value and fat echo factor are combined to determine the fat feature first evaluation value of enhanced analysis target block, and then determine the fat feature second evaluation value of each gray level, which helps to realize the enhancement of fat area differentiation under the condition that probe and wound skin contact is not good, and obtain enhanced ultrasound gray scale image for wound injury ultrasound evaluation with better enhancement effect.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical information processing, in particular to a wound condition ultrasonic evaluation method for free flap transplantation. BACKGROUND

[0002] Free flap transplantation is a complex microsurgical procedure used to repair large skin defects or tissue loss, and vascular crisis and flap necrosis after free flap transplantation also occur from time to time, and timely detection and active intervention can ensure the final survival of the flap. For free flap transplantation, postoperative wound condition evaluation focuses more on monitoring the survival of the flap after transplantation, blood supply, healing progress and complications, such as ultrasonic evaluation. Ultrasonic evaluation is a common preoperative wound condition evaluation method, which is non-invasive, real-time and easy to operate, helping doctors to understand the viability of the flap, blood flow, and whether there is edema, infection, etc.

[0003] When using a color Doppler ultrasound device to collect wound images, if the probe does not contact the wound well, the ultrasound images of different types of tissues may be affected to varying degrees, especially for fat and edema areas, both of which are prone to exhibit similar echo characteristics, i.e., both appear as low echo areas, especially on the wound, fat and edema or fluid accumulation areas may be difficult to distinguish on the image. Since the edema area may indicate that the wound after free flap transplantation has infection, inflammation, etc., therefore, there is an urgent need for a processing method to increase the difference in ultrasonic image features between the edema area and the fat. SUMMARY

[0004] In order to solve the technical problem that the fat and edema areas in the wound image are difficult to distinguish, resulting in low accuracy of the wound condition ultrasonic evaluation result, the purpose of the present application is to provide a wound condition ultrasonic evaluation method for free flap transplantation, and the technical solution adopted is as follows:

[0005] One embodiment of the present application provides a wound condition ultrasonic evaluation method for free flap transplantation, which comprises the following steps:

[0006] An ultrasonic gray-scale image of the wound area to be evaluated is obtained, and the ultrasonic gray-scale image is subjected to block processing to obtain each enhanced analysis candidate block;

[0007] Determine fat feature screening parameters according to the gray-scale distribution uniformity features and frequency shift data features of each enhanced analysis candidate block, and screen a plurality of enhanced analysis target blocks through the fat feature screening parameters;

[0008] According to the frequency shift data distribution features of each enhanced analysis target block, the blood vessel ultrasonic frequency shift feature value of each enhanced analysis target block is obtained;

[0009] calculate the fat echo factor of each enhanced analysis target sub-block, and determine a fat feature first evaluation value of each enhanced analysis target sub-block in combination with the blood vessel ultrasound frequency shift characteristic value;

[0010] determine a fat feature second evaluation value of each gray level according to the fat feature first evaluation value and the number of pixels corresponding to different gray levels in each enhanced analysis target sub-block;

[0011] In the histogram equalization process, the fat feature second evaluation value of each gray level is used to adjust the gray level mapping relationship, and then the ultrasound gray scale image is enhanced to obtain an enhanced ultrasound gray scale image.

[0012] Further, after obtaining the ultrasound gray scale image of the wound area to be evaluated, the method further comprises:

[0013] performing noise reduction processing on the ultrasound gray scale image of the wound area to be evaluated to obtain a noise-reduced ultrasound gray scale image.

[0014] Further, the determining of the fat feature screening parameter according to the gray level distribution uniformity feature and the frequency shift data feature of each enhanced analysis candidate sub-block comprises:

[0015] For any enhanced analysis candidate sub-block, the gray level information entropy and the gray level mean value are determined according to the gray level value of each pixel point in the enhanced analysis candidate sub-block, and then a first fat feature screening factor is determined;

[0016] wherein the gray level information entropy and the gray level mean value are negatively correlated with the first fat feature screening factor;

[0017] According to the frequency shift data of each pixel point in the enhanced analysis candidate sub-block, the average value of the absolute values of the frequency shift data of all pixel points is calculated to obtain a second fat feature screening factor;

[0018] wherein the average value is negatively correlated with the second fat feature screening factor; and the frequency shift data is the difference between the ultrasound wave transmission frequency and the echo frequency;

[0019] In combination with the first fat feature screening factor and the second fat feature screening factor, the fat feature screening parameter of the enhanced analysis candidate sub-block is determined.

[0020] Further, the obtaining of the blood vessel ultrasound frequency shift characteristic value of each enhanced analysis target sub-block according to the frequency shift data distribution feature of each enhanced analysis target sub-block comprises:

[0021] For any enhanced analysis target sub-block, the absolute values of the frequency shift data difference between each pixel point and all adjacent pixel points in the enhanced analysis target sub-block are calculated, and the minimum value of all absolute values corresponding to each pixel point is selected.

[0022] determine a first blood vessel ultrasonic frequency shift factor of the enhanced analysis target block according to the absolute value of the frequency shift data of each pixel point in the enhanced analysis target block, the minimum value corresponding to each pixel point, and the mean value of the frequency shift data of all pixel points;

[0023] determine the frequency shift eigenvalue of each pixel point in the enhanced analysis target block, and obtain each structure analysis pixel point in the enhanced analysis target block based on the frequency shift eigenvalue;

[0024] determine a second blood vessel ultrasonic frequency shift factor of the enhanced analysis target block according to the edge angle difference of each connected domain formed by the structure analysis pixel points;

[0025] determine the blood vessel ultrasonic frequency shift eigenvalue of the enhanced analysis target block in combination with the first blood vessel ultrasonic frequency shift factor and the second blood vessel ultrasonic frequency shift factor.

[0026] Further, the determination of the first blood vessel ultrasonic frequency shift factor of the enhanced analysis target block according to the absolute value of the frequency shift data of each pixel point in the enhanced analysis target block, the minimum value corresponding to each pixel point, and the mean value of the frequency shift data of all pixel points comprises:

[0027] for any pixel point in the enhanced analysis target block, calculate the difference value between the absolute value of the frequency shift data of the pixel point and the mean value of the frequency shift data; and obtain the frequency shift eigenvalue of the pixel point in combination with the difference value and the minimum value;

[0028] calculate the average value of the frequency shift eigenvalues of all pixel points in the enhanced analysis target block to obtain the first blood vessel ultrasonic frequency shift factor of the enhanced analysis target block.

[0029] Further, the determination of the second blood vessel ultrasonic frequency shift factor of the enhanced analysis target block according to the edge angle difference of each connected domain formed by the structure analysis pixel points comprises:

[0030] obtain the connected domain edge in any connected domain, segment the connected domain edge to obtain each edge segment, determine the edge principal axis of each edge segment, and obtain the included angle between each edge principal axis and the x-axis of the ultrasonic gray image;

[0031] calculate the difference value of the included angle between each edge principal axis and other edge principal axes in the connected domain except itself, and take the minimum included angle difference value corresponding to each edge principal axis as the characteristic direction value;

[0032] obtain the second blood vessel ultrasonic frequency shift factor of the enhanced analysis target block according to the characteristic direction value of each edge principal axis in each connected domain of the enhanced analysis target block.

[0033] Further, the calculating the fat echo factor of each enhanced analysis target sub-block comprises:

[0034] For any enhanced analysis target sub-block, a fitting edge curve of the enhanced analysis target sub-block is obtained, and then a curvature variance of each edge pixel point on the fitting edge curve is calculated;

[0035] According to a gray level variation feature of the pixel points in the enhanced analysis target sub-block extending to the edge, an effusion ultrasonic echo feature value of the enhanced analysis target sub-block is determined.

[0036] In combination with the curvature variance and the effusion ultrasonic echo feature value, a fat echo factor of the enhanced analysis target sub-block is determined.

[0037] The curvature variance and the effusion ultrasonic echo feature value are negatively correlated with the fat echo factor.

[0038] Further, the determining the effusion ultrasonic echo feature value of the enhanced analysis target sub-block according to the gray level variation feature of the pixel points in the enhanced analysis target sub-block extending to the edge comprises:

[0039] A geometric centroid coordinate point of the enhanced analysis target sub-block is obtained, and then a connecting line of each pixel point in the enhanced analysis target sub-block and the geometric centroid coordinate point is obtained, and a pixel point in a non-enhanced analysis target sub-block closest to the connecting line direction is taken as a contrast pixel point of a corresponding pixel point in the enhanced analysis target sub-block on the connecting line.

[0040] A distance value between each pixel point in the enhanced analysis target sub-block and the contrast pixel point thereof is calculated, and a gray level difference value between each pixel point in the enhanced analysis target sub-block and the geometric centroid coordinate point is calculated.

[0041] According to the distance value and the gray level difference value of each pixel point in the enhanced analysis target sub-block, the effusion ultrasonic echo feature value of the enhanced analysis target sub-block is determined.

[0042] The distance value is negatively correlated with the effusion ultrasonic echo feature value, and the gray level difference value is positively correlated with the effusion ultrasonic echo feature value.

[0043] Further, the determining the fat feature first evaluation value of each enhanced analysis target sub-block comprises:

[0044] For any enhanced analysis target sub-block, a blood vessel ultrasonic frequency shift feature value of the enhanced analysis target sub-block is negatively correlated to obtain a negative correlation value, and a product of the negative correlation value and the fat echo factor is taken as a fat feature first evaluation value of the enhanced analysis target sub-block.

[0045] Further, the method further comprises the following steps of:

[0046] For any one of the enhanced analysis target blocks, taking any one of the gray levels as a target gray level, determining a product of the fat feature first evaluation value and the number of pixel points of the target gray level in the enhanced analysis target block;

[0047] Calculating an accumulated value of the products corresponding to the target gray levels in all the enhanced analysis target blocks, and performing normalization processing on the accumulated value to obtain a normalized value, which is taken as the fat feature second evaluation value of the target gray level.

[0048] The present application has the following advantages:

[0049] The present application provides an ultrasound evaluation method for wound injury of free flap transplantation, which performs block processing on the obtained ultrasound gray scale image to obtain each enhanced analysis candidate block, and the block processing helps to improve the contrast between the fat area and the edema area. The method performs screening through fat feature screening parameters to obtain a plurality of enhanced analysis target blocks, which can be used to process the blocks that may meet the fat area feature after preliminary screening, thereby reducing the data analysis amount to a certain extent and improving the efficiency of the ultrasound evaluation of the wound injury. The method combines the blood vessel ultrasound frequency shift characteristic value and the fat echo factor to determine the fat feature first evaluation value of the enhanced analysis target block, and further determines the fat feature second evaluation value of each gray level, which helps to enhance the fat area differentiation under the condition that the probe does not contact the skin well. Compared with the prior art, the present application obtains an enhanced ultrasound gray scale image with better enhancement effect, which helps to improve the contrast between the fat area and the edema area, reduces the misjudgment probability of medical staff, reduces the complexity of the injury evaluation of the free flap transplantation wound under special circumstances, and further improves the patient experience of free flap transplantation. BRIEF DESCRIPTION OF DRAWINGS

[0050] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0051] Figure 1 The step flow chart of the ultrasound evaluation method for wound injury of free flap transplantation according to the embodiment of the present application is shown in the following figure.

[0052] Figure 2 The step flow chart of step S2 in the embodiment of the present application is shown in the following figure.

[0053] Figure 3 a step flow chart for step S3 in the embodiment of the present application;

[0054] Figure 4 a step flow chart for step S41 in the embodiment of the present application. DETAILED DESCRIPTION

[0055] In order to further clarify the technical means and effects taken by the present application to achieve the predetermined object, the specific implementation, structure, features and effects of the technical solutions according to the present application are described in detail below in combination with the drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0056] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0057] The scenario targeted by the present application is:

[0058] After free flap transplantation, wound condition assessment is an important part of postoperative monitoring, and ultrasound can provide non-invasive, real-time information to help assess the survival of the flap, blood supply, infection or complications, etc. When the postoperative wound is in a position that is difficult for some probes to reach (such as joint areas) or there is scab on the wound, the color Doppler ultrasound device probe will not be in good contact with the wound when collecting the ultrasound image of the wound, which will affect the clarity of the ultrasound image collected by the device through echo data, and further reduce the recognition of different internal tissue regions, resulting in low accuracy of the wound condition ultrasound assessment result.

[0059] In order to improve the feature difference between fat and edema regions in the wound ultrasound image and obtain a wound condition ultrasound assessment result with higher accuracy, the present embodiment provides a wound condition ultrasound assessment method for free flap transplantation, as shown in Figure 1 The method comprises the following steps:

[0060] S1, obtaining an ultrasound grayscale image of the wound area to be evaluated, and performing block processing on the ultrasound grayscale image to obtain each enhanced analysis candidate block.

[0061] The above step S1 can be realized by steps S11 to S12 (not shown in the figure):

[0062] S11, obtaining an ultrasound grayscale image of the wound area to be evaluated.

[0063] In this embodiment, an ultrasound device is used to collect an ultrasound image of a wound surface area to be evaluated after free flap transplantation. The success of free flap transplantation is closely related to the blood supply of the flap, so a color Doppler ultrasound device is selected to collect a color Doppler ultrasound image of the wound surface area to be evaluated, which is subjected to grayscale processing to obtain an ultrasound grayscale image.

[0064] Among them, the color Doppler blood flow imaging technology is to display the blood flow signal in color, and the pseudo-color coding is composed of three basic colors of red, blue and green. For example, red represents blood flow towards the probe, blue represents blood flow away from the probe, and in addition, the blood flow velocity is related to the color brightness. The higher the velocity, the stronger the color brightness, and the lower the velocity, the weaker the color brightness. The color Doppler ultrasound image usually reflects the above pseudo-color process in a combination of grayscale image and color coding. The original data collected is usually a grayscale image and frequency shift information, that is, for an ultrasound grayscale image, in addition to its own grayscale value, it also carries frequency shift data. The grayscale image is used to display the tissue structure, and the grayscale can reflect the intensity of the reflection echo signal of the corresponding pixel point. The frequency shift information is used to calculate the velocity and direction of the blood flow, and the frequency shift specifically refers to the difference between the frequency of the ultrasonic wave emitted and the frequency of the echo.

[0065] After obtaining the ultrasound grayscale image of the wound surface area to be evaluated, the following steps are further included:

[0066] The ultrasound grayscale image of the wound surface area to be evaluated is subjected to noise reduction processing to obtain a noise-reduced ultrasound grayscale image.

[0067] In this embodiment, the ultrasound grayscale image is subjected to noise reduction processing by using bilateral filtering, which can smooth the noise while retaining the edge texture features in the grayscale image. The implementation process of bilateral filtering for noise reduction processing is prior art and is not within the protection scope of the present application, and will not be described in detail here.

[0068] It is worth noting that in the subsequent steps, if not specifically stated, the ultrasound grayscale image with superimposed frequency shift data is operated.

[0069] It should be noted that when the wound surface after free flap transplantation is scabbed or in a position where the probe is difficult to reach (such as joint parts, etc.), the ultrasound probe may not be able to fully contact the skin of the wound surface during detection. Poor contact between the ultrasound probe and the wound surface can reduce the efficiency of sound wave transmission, resulting in a decrease in the ultrasound echo signal in some areas. Especially for fat, due to its low echo, in the case of poor contact between the probe and the skin of the wound surface, the image of the boundary between fat and surrounding tissue is less clear, which may be confused with the edema area caused by infection, affecting the accuracy of the ultrasound evaluation of the wound surface area after free flap transplantation. Therefore, the image features of the fat area need to be enhanced to ensure the evaluation effect of the ultrasound evaluation of the wound surface injury recovery.

[0070] S12, the ultrasound grayscale image is segmented to obtain various candidate blocks for enhancement analysis.

[0071] Enhancement processing of regions representing fat in ultrasound grayscale images aims to improve the differentiation between fat and edema. During normal acquisition, both fat and edema regions appear as relatively simple areas with simple boundary shapes in ultrasound grayscale images, rather than complex shapes such as nerves or blood vessels. To improve the contrast between fat and edema regions, the ultrasound grayscale image needs to be segmented. Subsequently, feature analysis of the segmented images can be used to select target segments that may be fat, and then enhancement processing can be performed on these segments based on their image features.

[0072] In this embodiment, a superpixel segmentation algorithm is used to process the ultrasound grayscale image, dividing the image into multiple superpixel blocks, i.e., candidate blocks for enhancement analysis. Specifically, the SLIC algorithm (Simple Linear Iterative Clustering) is used to perform superpixel segmentation on the ultrasound grayscale image. The number of superpixel blocks (candidate blocks for enhancement analysis) can be obtained using the elbow method. Candidate blocks for enhancement analysis consist of pixels with similar texture and grayscale features. The implementation processes of the SLIC algorithm and the elbow method are existing technologies and are not within the scope of this invention; therefore, they will not be described in detail here.

[0073] Thus, this embodiment has obtained each candidate block for enhancement analysis corresponding to the ultrasonic grayscale image.

[0074] S2, determine the fat feature screening parameters based on the gray-level distribution uniformity and frequency shift data characteristics of each enhancement analysis candidate block, and obtain several enhancement analysis target blocks by screening through the fat feature screening parameters.

[0075] It should be noted that, in order to enhance the fat image features in ultrasound grayscale images, candidate blocks for enhancement analysis were screened. Ultrasound grayscale images typically contain numerous subcutaneous tissues. Since fat regions usually do not contain other tissues or organs, they are relatively uniform within the block area of ​​the ultrasound grayscale image; the echo intensity within fat regions is usually low, resulting in smaller grayscale values ​​for fat regions in ultrasound images.

[0076] The above step S2 can be achieved through Figure 2 Steps S21 to S24 shown are implemented as follows:

[0077] S21, for any candidate block for enhancement analysis, determine the gray information entropy and gray mean value based on the gray value of each pixel in the candidate block for enhancement analysis, and then determine the first fat feature screening factor.

[0078] Here, the first fat feature screening factor can represent the echo feature of the enhanced analysis candidate sub-block, i.e., the gray scale distribution uniformity feature.

[0079] In this embodiment, the gray scale information entropy and the gray scale mean value are negatively correlated with the first fat feature screening factor, and the case where the gray scale information entropy or the gray scale mean value is zero is not analyzed.

[0080] As an example, the product of the gray scale information entropy and the gray scale mean value is calculated first, and then the reciprocal of the product is determined as the first fat feature screening factor of the enhanced analysis candidate sub-block.

[0081] Since the gray scale distribution of the fat region is relatively concentrated, and the gray scale value is relatively low, the smaller the product of the gray scale information entropy and the gray scale mean value, the larger the first fat feature screening factor, indicating that the fat feature contained in the enhanced analysis candidate sub-block is more obvious.

[0082] S22, according to the frequency shift data of each pixel point in the enhanced analysis candidate sub-block, the average value of the absolute values of the frequency shift data of all pixel points is calculated to obtain a second fat feature screening factor.

[0083] Here, the second fat feature screening factor represents the frequency shift data feature.

[0084] In this embodiment, the average value of the absolute values of the frequency shift data of all pixel points in the enhanced analysis candidate sub-block is negatively correlated with the second fat feature screening factor.

[0085] As an example, the reciprocal of the average value of the absolute values of the frequency shift data of all pixel points is taken as the second fat feature screening factor of the enhanced analysis candidate sub-block, and generally the average value of the absolute values of the frequency shift data of all pixel points does not have the possibility of being zero.

[0086] Since the fat region does not have blood vessel distribution, it will not show obvious flow features in the color Doppler ultrasound image, i.e., the frequency shift data carried by the pixel points in the fat region is usually zero, therefore, the smaller the average value of the absolute values of the frequency shift data of all pixel points in the enhanced analysis candidate sub-block, the larger the second fat feature screening factor.

[0087] S23, the first fat feature screening factor and the second fat feature screening factor are combined to determine the fat feature screening parameter of the enhanced analysis candidate sub-block.

[0088] In this embodiment, the echo feature and the frequency shift data feature of the enhanced analysis candidate sub-block are combined, i.e., the product of the first fat feature screening factor and the second fat feature screening factor is taken as the fat feature screening parameter of the enhanced analysis candidate sub-block, and the larger the first fat feature screening factor and the second fat feature screening factor, the more obvious the echo feature and the frequency shift data feature of the enhanced analysis candidate sub-block.

[0089] S24, screening is performed on the selected sub-blocks by using the fat feature screening parameters of the enhanced analysis, to obtain a plurality of enhanced analysis target sub-blocks.

[0090] As an example, referring to the calculation method of the fat feature screening parameters of the enhanced analysis selected sub-blocks in any one of the above steps S21 to S23, the fat feature screening parameters of the enhanced analysis selected sub-blocks can be obtained; then, all the fat feature screening parameters are arranged in descending order to obtain a screening parameter sequence; next, the difference between adjacent elements in the screening parameter sequence is calculated, and the larger screening parameter of the two elements corresponding to the maximum difference is taken as the target screening parameter; finally, the enhanced analysis selected sub-blocks greater than or equal to the target screening parameter in the screening parameter sequence are taken as the enhanced analysis target sub-blocks.

[0091] It is worth noting that by calculating the difference between adjacent elements in the screening parameter sequence and selecting the two elements with the maximum difference, the region with significant feature change can be effectively identified; taking the larger screening parameter of the two elements corresponding to the maximum difference as the target screening parameter helps to take the enhanced analysis selected sub-blocks with obvious feature parameter change as the analysis target, which can improve the analysis accuracy because it focuses on the region with the most significant feature change; selectively selecting the enhanced analysis target sub-blocks can more effectively utilize resources and focus on the part that is most likely to provide valuable information.

[0092] At this point, the embodiment obtains a plurality of enhanced analysis target sub-blocks corresponding to the ultrasound grayscale image.

[0093] It should be noted that the enhanced analysis target sub-blocks are initially screened by comparing the regional overall features of the superpixel blocks with other regions in the ultrasound grayscale image, but in the case of poor contact between the device probe and the surface of the creation area, it is still difficult to distinguish the superpixel blocks of the edema region and the fat region after the initial screening. Therefore, in order to perform enhanced processing on the fat region, further analysis is needed, i.e., steps S3 to S5.

[0094] S3, according to the frequency shift data distribution characteristics of each enhanced analysis target sub-block, the blood vessel ultrasound frequency shift feature value of each enhanced analysis target sub-block is obtained.

[0095] It should be noted that in Doppler ultrasound, frequency shift data is obtained by the frequency change caused by the relative movement between ultrasonic waves and moving objects (such as red blood cells). When the ultrasonic probe and the surface of the wound area are not in good contact, it may cause abnormal frequency shift data. In detail, in the actual use of the probe, the abnormal change of the frequency shift data caused by the poor contact between the ultrasonic probe and the surface of the wound area should have similar effects on all pixel points in the same superpixel block in the ultrasonic gray scale image, because the surface of the ultrasonic probe is usually flat, and when the probe is not in good contact with the surface of the wound area, the difference between the distances of each receiving crystal on the probe to different positions of the tissue corresponding to the same superpixel block is still small. Specifically, for the superpixel block of fat not containing blood vessels, the characteristic of the abnormal change of the frequency shift data should be that there is a small difference; and for the edema area, whether the blood vessels in the edema area depend on the specific swelling area, even if blocked by the edema area, the frequency shift data generated by the blood vessels is also larger than that of other areas. Therefore, the fat area and the edema area can be further distinguished by analyzing the difference in the frequency shift data carried by the pixel points in the enhanced analysis target block.

[0096] In addition, in the process of analyzing the frequency shift data, in order to further reduce the influence of the poor contact between the ultrasonic probe and the surface of the wound area, the structural characteristics of the blood vessels need to be analyzed: for the pixel points representing the frequency shift data of the blood vessels, the surrounding usually has pixel points capable of representing the frequency shift data of the blood vessels, and the pixel points with the blood vessel characteristics are amplified by the distribution characteristics of the pixel points with the blood vessel frequency shift data, and the subsequent calculation process is adjusted.

[0097] For any enhanced analysis target block, the above step S3 can be implemented by steps S31 to S35 as shown in the following table: Figure 3

[0098] S31, calculate the absolute value of the frequency shift data difference between each pixel point in the enhanced analysis target block and all its adjacent pixel points, and select the minimum value of all the absolute values corresponding to each pixel point.

[0099] In this embodiment, the pixel points adjacent to the pixel points in the enhanced analysis target block are usually the pixel points in the eight adjacent domains, and each pixel point in the enhanced analysis target block corresponds to at least one adjacent pixel point, and then the minimum value is selected from the absolute value of the frequency shift data difference between the pixel point and its adjacent pixel point, and each pixel point has its corresponding minimum value.

[0100] S32, according to the frequency shift data absolute value of each pixel point in the enhanced analysis target block, the minimum value corresponding to each pixel point, and the average value of the frequency shift data of all pixel points, determine the first blood vessel ultrasonic frequency shift factor of the enhanced analysis target block.

[0101] ​Here, the first vascular ultrasound frequency shift factor represents the degree of characteristic performance of the blood vessels blocked by the edema through the frequency shift data distribution analysis.

[0102] The above step S32 can be implemented by steps S321 to S322 (not shown in the figure):

[0103] S321, for any pixel point in the enhanced analysis target block, calculate the difference value between the absolute value of the frequency shift data of the pixel point and the mean value of the frequency shift data; obtain the frequency shift characteristic value of the pixel point in combination with the difference value and the minimum value.

[0104] S322, calculate the average value of the frequency shift characteristic values of all pixel points in the enhanced analysis target block, and obtain the first vascular ultrasound frequency shift factor of the enhanced analysis target block.

[0105] As an example, the calculation formula of the first vascular ultrasound frequency shift factor of the i-th enhanced analysis target block can be:

[0106] In the formula, represents the first vascular ultrasound frequency shift factor of the i-th enhanced analysis target block, i represents the i-th enhanced analysis target block, n i represents the number of pixel points in the i-th enhanced analysis target block, j represents the j-th pixel point in the i-th enhanced analysis target block, represents the absolute value of the frequency shift data of the j-th pixel point in the i-th enhanced analysis target block, represents the mean value of the frequency shift data of all pixel points in the i-th enhanced analysis target block, represents the minimum value of the j-th pixel point in the i-th enhanced analysis target block, represents the frequency shift characteristic value of the j-th pixel point in the i-th enhanced analysis target block.

[0107] In the calculation formula of the first vascular ultrasound frequency shift factor, represents the distance between the frequency shift data of a single pixel point in the enhanced analysis target block and the frequency shift data of the whole block, The larger it is, the more the j-th pixel point deviates from the frequency shift size of the whole region, the more likely the j-th pixel point is a blood vessel pixel point covered by the edema region, and the larger the first vascular ultrasound frequency shift factor is. The blood vessel pixel point is usually surrounded by other pixel points that can represent the frequency shift data characteristics of the blood vessel, The smaller it is, the smaller the frequency shift data difference between the j-th pixel point and its surrounding pixel points, the more likely it is to conform to the blood vessel structure characteristics, and the larger the first vascular ultrasound frequency shift factor will be.

[0108] S33, determine the frequency shift characteristic value of each pixel point in the enhanced analysis target block, and obtain each structure analysis pixel point in the enhanced analysis target block based on the frequency shift characteristic value.

[0109] It should be noted that the blood vessels are usually distributed in the form of a straight or curved tube, so after quantifying the blood vessel frequency shift feature of each pixel point in the enhanced analysis target block, the pixel points with higher obviousness are analyzed in the ultrasound gray scale image to further improve the accuracy of data analysis. First, select the pixel points with required blood vessel structure features in the enhanced analysis target block, denoted as structure analysis pixel points.

[0110] In this embodiment, the frequency shift characteristic value of each pixel point in the enhanced analysis target block is obtained based on the calculation method of the frequency shift characteristic value in step S32; the frequency shift characteristic values of the same enhanced analysis target block are arranged in descending order to obtain a frequency shift characteristic value sequence; the difference between adjacent elements in the frequency shift characteristic value sequence is calculated, and then the two elements with the largest difference are selected, and the pixel points with a frequency shift characteristic value greater than or equal to the larger value of the two selected frequency shift characteristic values are selected as structure analysis pixel points.

[0111] S34, determine the second blood vessel ultrasound frequency shift factor of the enhanced analysis target block according to the edge angle difference of the edges of each connected domain composed of the structure analysis pixel points.

[0112] The above step S34 can be realized by steps S341 to S343 (not shown in the figure):

[0113] S341, obtain the connected domain edges in any connected domain, segment the connected domain edges to obtain each edge segment, determine the edge principal axis of each edge segment, and obtain the included angle between each edge principal axis and the x-axis of the ultrasound gray scale image.

[0114] In this embodiment, the structure analysis pixel points of the current enhanced analysis target block are detected to divide them into multiple connected domains, which helps to filter out single discrete pixel points; for the edge of the tubular structure, another side can usually be found to be approximately parallel to its direction, so the angle difference between the edges of the same connected domain (the principal axis edges of the edge segments) can be evaluated to analyze the blood vessel features exhibited by the connected domain to determine the second blood vessel ultrasound frequency shift factor.

[0115] Specifically, for any enhanced analysis target sub-block, the connected domains are obtained based on the pixels in the enhanced analysis target sub-block, the edges of the connected domains are obtained from the connected domains, the edge segments are obtained by segmenting the edges of the connected domains, and the edge principal axes of the edge segments are calculated by using the PCA (Principal Component Analysis). The x-axis of the ultrasound grayscale image, i.e., the horizontal direction, is obtained, and the angle between the edge principal axes and the x-axis of the ultrasound grayscale image is obtained.

[0116] For example, for the edges of the connected domains, the implementation process of segmenting the edges of the connected domains to obtain the edge segments can include the following steps.

[0117] The slope difference between every two adjacent edge pixels on the edge of the connected domain is calculated, the slope difference is sorted from large to small, the larger slope corresponding to the two edge pixels with the largest slope difference in the sequence is selected as a reference value, the edge pixels on the edge of the connected domain greater than or equal to the reference value are selected as the segmentation points, and the edge segments of the edge of the connected domain are obtained by segmenting the edge of the connected domain.

[0118] S342, the difference value of the angle between the edge principal axis and other edge principal axes in the connected domain except itself is calculated, and the minimum difference value of the angle corresponding to each edge principal axis is taken as the characteristic direction value.

[0119] In this embodiment, for any edge principal axis, the difference value of the angle between the edge principal axis and other edge principal axes in the connected domain except itself is determined, and each difference value corresponding to the edge principal axis is obtained, and the minimum difference value is taken as the characteristic direction value of the corresponding edge principal axis.

[0120] S343, the second blood vessel ultrasound frequency shift factor of the enhanced analysis target sub-block is obtained according to the characteristic direction values of the edge principal axes in each connected domain of the enhanced analysis target sub-block.

[0121] In this embodiment, the average value of the characteristic direction values of all edge principal axes in the same connected domain is calculated first, and then the average value of the average values of the characteristic direction values in all connected domains in the same enhanced analysis target sub-block is calculated, which is taken as the second blood vessel ultrasound frequency shift factor of the enhanced analysis target sub-block. The greater the second blood vessel ultrasound frequency shift factor is, the more significant the blood vessel structure feature represented by the frequency shift data carried by the enhanced analysis target sub-block is.

[0122] S35, the blood vessel ultrasound frequency shift characteristic value of the enhanced analysis target sub-block is determined by combining the first blood vessel ultrasound frequency shift factor and the second blood vessel ultrasound frequency shift factor.

[0123] In this embodiment, the frequency shift data difference of the enhanced analysis target block is combined with the structure analysis of the frequency shift data structure characteristics carried by the pixel points to determine the vascular ultrasound frequency shift characteristic value of the enhanced analysis target block.

[0124] As an example, the product of the first vascular ultrasound frequency shift factor and the second vascular ultrasound frequency shift factor of the same enhanced analysis target block is taken as the vascular ultrasound frequency shift characteristic value, and the larger the vascular ultrasound frequency shift characteristic value, the more likely it is that the frequency shift data carried by the pixel points in the enhanced analysis target block has the vascular ultrasound feature. Referring to the implementation steps of the vascular ultrasound frequency shift characteristic value of any one of the enhanced analysis target blocks, the vascular ultrasound frequency shift characteristic value of each enhanced analysis target block can be obtained.

[0125] Up to now, the vascular ultrasound frequency shift characteristic value of each enhanced analysis target block has been obtained in this embodiment.

[0126] S4, the fat echo factor of each enhanced analysis target block is calculated, and the vascular ultrasound frequency shift characteristic value is combined to determine the fat feature first evaluation value of each enhanced analysis target block.

[0127] Here, the fat echo factor refers to the degree of compliance with the echo intensity characteristics of the fat region, and the fat feature first evaluation value refers to the degree of compliance of the enhanced analysis target block with the image characteristics of the fat region. In addition to analyzing the frequency shift data, i.e., determining the vascular ultrasound frequency shift characteristic value, further evaluation is needed through the echo intensity characteristic difference between the fat region and the edema region in this embodiment.

[0128] The above step S4 can be implemented by steps S41 to S42 (not shown in the figure):

[0129] S41, the fat echo factor of each enhanced analysis target block is calculated.

[0130] For any enhanced analysis target block, the above step S41 can be implemented by steps S411 to S413 shown in the figure: Figure 4

[0131] S411, the fitting edge curve of the enhanced analysis target block is obtained, and the curvature variance of each edge pixel point on the fitting edge curve is calculated.

[0132] It should be noted that the edema region is often in the form of effusion in human tissue and often does not have a regular structure in the ultrasound image. However, the fat region is composed of fat cells, which has a certain structural stability compared with the edema region, so the edge shape feature of the enhanced analysis target block can be used as one of the factors for evaluating the fat echo factor.

[0133] ​Specifically, the least square method is used to obtain a fitting edge curve of the enhanced analysis target sub-block, and then a variance of curvature of a fitting point corresponding to each edge pixel point on the edge of the enhanced analysis target sub-block on the fitting edge curve is calculated. The greater the curvature variance is, the less likely the edge of the enhanced analysis target sub-block has a regular structure feature, the greater the possibility of being an edema region is, and the less the degree of conforming to the echo intensity feature of a fat region is. The least square method and the calculation process of the curvature are both prior art and are not within the protection scope of the present application, and thus are not described in detail here.

[0134] S412, determining an effusion ultrasound echo feature value of the enhanced analysis target sub-block according to a gray scale variation feature of the pixel point in the enhanced analysis target sub-block to the edge.

[0135] It should be noted that in addition to the edge shape difference caused by different forms, the effusion of the edema region often has an expansion feature, that is, has a feature that the effusion flows from the region center to the edge, causing the ultrasound signal originally blocked by the effusion to gradually weaken, and the background tissue reflection ability to become stronger and stronger; and the fat region usually has a uniform echo performance and a clear boundary. Therefore, the gray scale variation feature of different sub-block pixel points to the edge can be used as another factor for evaluating the fat echo factor.

[0136] The above step S412 can be implemented through steps S4121 to S4123 (not shown in the figure):

[0137] S4121, obtaining a geometric centroid coordinate point of the enhanced analysis target sub-block, and then obtaining a connecting line between each pixel point in the enhanced analysis target sub-block and the geometric centroid coordinate point, and taking a pixel point in a non-enhanced analysis target sub-block closest to the connecting line direction as a contrast pixel point corresponding to the pixel point in the enhanced analysis target sub-block on the connecting line.

[0138] In the embodiment, the pixel point in the non-enhanced analysis target sub-block refers to a pixel point in the ultrasound gray scale image that is not located in the enhanced analysis target sub-block, and the contrast pixel point is obtained to analyze the feature of flowing from the center to the edge. Each pixel point in the enhanced analysis target sub-block has its corresponding contrast pixel point. It should be noted that the pixel points in the enhanced analysis target sub-block located in the same connecting line direction have the same contrast pixel point.

[0139] S4122, calculating a distance value between each pixel point in the enhanced analysis target sub-block and its contrast pixel point, and calculating a gray scale difference value between each pixel point in the enhanced analysis target sub-block and the geometric centroid coordinate point.

[0140] In this embodiment, the distance value between the pixel point and the contrast pixel point can be an Euclidean distance value, which can be used as an adjustment parameter of the gray value difference between the pixel point and the geometric centroid coordinate point. The smaller the distance value, the greater the possibility that the corresponding pixel point in the enhanced analysis target block is located at the edge of the block, and the greater the degree of influence of the effusion ultrasound echo.

[0141] In this embodiment, the gray value difference refers to the difference between the gray value of the pixel point in the enhanced analysis target block and the gray value of the geometric centroid coordinate point. The greater the gray value difference, the stronger the background tissue reflection ability of the pixel point, and the more significant the effusion ultrasound echo feature of the pixel point in the enhanced analysis target block.

[0142] S4123, according to the distance value and the gray value difference of each pixel point in the enhanced analysis target block, determine the effusion ultrasound echo feature value of the enhanced analysis target block.

[0143] In this embodiment, the distance value is negatively correlated with the effusion ultrasound echo feature value, while the gray value difference is positively correlated with the effusion ultrasound echo feature value.

[0144] As an example, the calculation formula of the effusion ultrasound echo feature value of the i-th enhanced analysis target block can be:

[0145] E i denotes the effusion ultrasound echo feature value of the i-th enhanced analysis target block, sigmoid denotes a normalization function, i.e. a nonlinear activation function, n i denotes the number of pixel points in the i-th enhanced analysis target block, j denotes the j-th pixel point in the i-th enhanced analysis target block, d ij denotes the distance value between the j-th pixel point in the i-th enhanced analysis target block and the contrast pixel point of the j-th pixel point, Δg ij denotes the gray value difference between the j-th pixel point in the i-th enhanced analysis target block and the geometric centroid coordinate point.

[0146] In the calculation formula of the effusion ultrasound echo feature value, the smaller the distance between the pixel point and the corresponding non-enhanced analysis target block pixel point, and the greater the gray value difference between the pixel point and the geometric centroid coordinate point, the more significant the effusion ultrasound echo feature presented by the enhanced analysis target block, and the greater the effusion ultrasound echo feature value; since Δg ij There is a possibility of being negative, so the value range of the effusion ultrasound echo feature value is limited to 0 to 1 by the sigmoid function. The greater the effusion ultrasound echo feature value, the more likely the enhanced analysis target block has the echo feature of the effusion region.

[0147] S413, determine the fat echo factor of the enhanced analysis target sub-block in combination with the curvature variance and the effusion ultrasound echo feature value.

[0148] In this embodiment, both the curvature variance and the effusion ultrasound echo feature value are negatively correlated with the fat echo factor.

[0149] As an example, the calculation formula of the fat echo factor of the i-th enhanced analysis target sub-block can be:

[0150] In the formula, K i represents the fat echo factor of the i-th enhanced analysis target sub-block, represents the curvature variance of the i-th enhanced analysis target sub-block, E i represents the effusion ultrasound echo feature value of the i-th enhanced analysis target sub-block, and +0.1 is to avoid the case that the denominator of the fraction is zero. Of course, 0.1 can be replaced by other non-zero constants, which are not specifically limited here.

[0151] In the calculation formula of the fat echo factor, the greater the curvature variance, the worse the structural regularity of the enhanced analysis target sub-block, and the less the fat echo factor of the enhanced analysis target sub-block conforms to the stability characteristics of the fat region. The greater the effusion ultrasound echo feature value, the more the enhanced analysis target sub-block conforms to the effusion expansion characteristics of the edema region, and the less the enhanced analysis target sub-block conforms to the relatively uniform echo performance of the fat region, so the fat echo factor of the enhanced analysis target sub-block is smaller.

[0152] S42, determine the fat feature first evaluation value of each enhanced analysis target sub-block according to the fat echo factor and the blood vessel ultrasound frequency shift feature value of each enhanced analysis target sub-block.

[0153] In this embodiment, the blood vessel ultrasound frequency shift feature and the fat echo factor are combined to evaluate the fat region probability of the enhanced analysis target sub-block, that is, to determine the fat feature first evaluation value of the enhanced analysis target sub-block. The blood vessel ultrasound frequency shift feature value is negatively correlated with the fat feature first evaluation value, and the fat echo factor is positively correlated with the fat feature first evaluation value.

[0154] Specifically, for any enhanced analysis target sub-block, the blood vessel ultrasound frequency shift feature value of the enhanced analysis target sub-block is negatively correlated to obtain a negative correlation value, and the product of the negative correlation value and the fat echo factor is taken as the fat feature first evaluation value of the enhanced analysis target sub-block. In this embodiment, the means of negative correlation processing is not specifically limited, and the implementer can set it according to the specific actual situation, such as taking the reciprocal of the blood vessel ultrasound frequency shift feature value, or using an exponential function exp(-) with a natural constant as the base to realize it.

[0155] As an example, the calculation formula of the fat feature first evaluation value of the i-th enhanced analysis target sub-block can be:

[0156] In the formula, Q i represents the first evaluation value of the fat feature of the i-th enhanced analysis target sub-block, γ i represents the first evaluation value of the fat feature of the i-th enhanced analysis target sub-block, +0.1 is to avoid the case that the denominator of the fraction is zero, of course, 0.1 can also be replaced by other non-zero constants, which is not limited here.

[0157] So far, the first evaluation value of the fat feature of each enhanced analysis target sub-block in the ultrasound grayscale image has been obtained.

[0158] S5, according to the first evaluation value of the fat feature and the number of pixels corresponding to each gray level in each enhanced analysis target sub-block, the second evaluation value of the fat feature of each gray level is determined.

[0159] It should be noted that after obtaining the first evaluation value of the fat feature of each enhanced analysis target sub-block, the significant degree of the fat feature of different gray levels is evaluated and analyzed in combination with the first evaluation value of the fat feature of the sub-block, that is, the second evaluation value of the fat feature of each gray level in the enhanced analysis target sub-block is determined.

[0160] In this embodiment, the ultrasound echo of the fat region is relatively uniform, so the significant degree of the fat feature of different gray levels can be evaluated in combination with the first evaluation value of the fat feature of different sub-blocks and the number of pixel points of different gray levels in the enhanced analysis target sub-block.

[0161] Specifically, for any enhanced analysis target sub-block, any gray level is taken as a target gray level, and the product of the first evaluation value of the fat feature and the number of pixel points of the target gray level in the enhanced analysis target sub-block is determined; the accumulation value of the products corresponding to the target gray level in all enhanced analysis target sub-blocks is calculated, and the accumulation value is normalized to obtain a normalized value, which is taken as the second evaluation value of the fat feature of the target gray level.

[0162] As an example, the calculation formula of the second evaluation value of the fat feature of the m-th gray level can be:

[0163] In the formula, P m represents the second evaluation value of the fat feature of the m-th gray level, norm represents the range normalization function, U represents the number of enhanced analysis target sub-blocks in the ultrasound grayscale image, Q i represents the first evaluation value of the fat feature of the i-th enhanced analysis target sub-block, num mi represents the number of pixel points of the m-th gray level in the i-th enhanced analysis target sub-block.

[0164] In the calculation formula of the second evaluation value of the fat feature, the first evaluation value of the fat feature of different gray scales in the same enhancement analysis target block is the same, but the fat feature prominence of the same gray scale in different enhancement analysis target blocks can be different, and when the fat feature prominence of the same gray scale in different enhancement analysis target blocks is large, the fat feature prominence of the gray scale is more obvious. Referring to the calculation method of the second evaluation value of the fat feature of the mth gray scale, the second evaluation value of the fat feature of each gray scale can be obtained.

[0165] So far, the second evaluation value of the fat feature of each gray scale corresponding to the enhancement analysis target block is obtained in the embodiment.

[0166] S6, in the histogram equalization process, the second evaluation value of the fat feature of each gray scale is used to adjust the gray scale mapping relationship, and then the ultrasound gray scale image is enhanced to obtain an enhanced ultrasound gray scale image.

[0167] In the embodiment, after obtaining the second evaluation value of the fat feature of each gray scale corresponding to the enhancement analysis target block, the image is enhanced by histogram equalization, and the gray scale mapping relationship is adjusted.

[0168] As an example, the calculation formula of the adjusted gray scale mapping relationship can be:

[0169] In the formula, s k represents the adjusted gray scale mapping relationship in the histogram equalization process, round represents the rounding function, L represents the number of gray scales, k represents the kth gray scale, P m represents the second evaluation value of the fat feature of the mth gray scale in the ultrasound gray scale image, P a represents the cumulative value of the second evaluation value of the fat feature of all gray scales in the ultrasound gray scale image, n m represents the number of pixel points of the mth gray scale in the ultrasound gray scale image, and N represents the number of all pixel points in the ultrasound gray scale image.

[0170] It should be noted that there can be a case where the number of gray scales of the ultrasound gray scale image and the number of gray scales corresponding to the enhancement analysis target block are not equal, in which case the embodiment sets the gray scale not appearing on the enhancement analysis target block to 1.

[0171] ​Finally, the ultrasound gray scale image is enhanced by combining the adjusted gray scale mapping relationship and the CLAHE (Contrast Limited Adaptive Histogram Equalization) method to obtain an enhanced ultrasound gray scale image, which is helpful to enhance the fat differentiation in the case of poor contact between the probe and the skin of the wound and to improve the differentiation between the fat region and the edema tissue. The implementation process of the CLAHE method is prior art and is not within the protection scope of the present application, and thus is not described in detail here.

[0172] After obtaining the enhanced ultrasound gray scale image, the image is pseudo-color processed by using the frequency shift data. The pseudo-color processing is prior art and is not described in detail here. The image after the pseudo-color processing is output.

[0173] In summary, the present application increases the differentiation between the fat region and the edema region by using the enhanced ultrasound gray scale image, improves the reliability of the ultrasound image collected in the wound condition assessment process, and provides an effective basis for the diagnosis of the wound condition by medical personnel.

[0174] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can be modified, or some technical features can be replaced by equivalent ones; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. An ultrasonic method for assessing the wound condition of a free flap graft, characterized in that, The method comprises the following steps: obtaining an ultrasound grayscale image of a wound area to be evaluated, and performing block processing on the ultrasound grayscale image to obtain each enhanced analysis candidate block; determining fat feature screening parameters according to the grayscale distribution uniformity feature and the frequency shift data feature of each enhanced analysis candidate block, and screening to obtain a plurality of enhanced analysis target blocks through the fat feature screening parameters; obtaining a blood vessel ultrasound frequency shift feature value of each enhanced analysis target block according to the frequency shift data distribution feature of each enhanced analysis target block; calculating a fat echo factor of each enhanced analysis target block, and determining a fat feature first evaluation value of each enhanced analysis target block in combination with the blood vessel ultrasound frequency shift feature value; determining a fat feature second evaluation value of each gray level according to the fat feature first evaluation value and the number of pixels corresponding to different gray levels in each enhanced analysis target block; in a histogram equalization process, adjusting a gray level mapping relationship by using the fat feature second evaluation value of each gray level, and then performing image enhancement on the ultrasound grayscale image to obtain an enhanced ultrasound grayscale image; the method of obtaining a blood vessel ultrasound frequency shift feature value of each enhanced analysis target block according to the frequency shift data distribution feature of each enhanced analysis target block comprises: for any enhanced analysis target block, calculating the absolute value of the frequency shift data difference value between each pixel point and all adjacent pixel points in the enhanced analysis target block, and selecting the minimum value of all absolute values corresponding to each pixel point; determining a first blood vessel ultrasound frequency shift factor of the enhanced analysis target block according to the frequency shift data absolute value of each pixel point in the enhanced analysis target block, the minimum value corresponding to each pixel point, and the frequency shift data mean value of all pixel points; determining a frequency shift feature value of each pixel point in the enhanced analysis target block based on the frequency shift feature value, and obtaining each structure analysis pixel point in the enhanced analysis target block; determining a second blood vessel ultrasound frequency shift factor of the enhanced analysis target block according to the edge angle difference of the edge of each connected domain formed by the structure analysis pixel points; combining the first blood vessel ultrasound frequency shift factor and the second blood vessel ultrasound frequency shift factor to determine the blood vessel ultrasound frequency shift feature value of the enhanced analysis target block; the method of determining a first blood vessel ultrasound frequency shift factor of the enhanced analysis target block according to the frequency shift data absolute value of each pixel point in the enhanced analysis target block, the minimum value corresponding to each pixel point, and the frequency shift data mean value of all pixel points comprises: for any pixel point in the enhanced analysis target block, calculating the difference value between the frequency shift data absolute value of the pixel point and the frequency shift data mean value; combining the difference value and the minimum value to obtain the frequency shift feature value of the pixel point; calculating the average value of the frequency shift feature values of all pixel points in the enhanced analysis target block to obtain the first blood vessel ultrasound frequency shift factor of the enhanced analysis target block; the method of determining a second blood vessel ultrasound frequency shift factor of the enhanced analysis target block according to the edge angle difference of the edge of each connected domain formed by the structure analysis pixel points comprises: Obtaining a connected domain edge in any connected domain, performing segmentation on the connected domain edge to obtain each edge segment, determining an edge principal axis of each edge segment, and obtaining an included angle between each edge principal axis and an x-axis of the ultrasound grayscale image; Calculating a difference value of the included angle between each edge principal axis and other edge principal axes in the connected domain except itself, and taking a minimum included angle difference value corresponding to each edge principal axis as a characteristic direction value; According to the characteristic direction values of each edge principal axis in each connected domain of the enhanced analysis target block, a second blood vessel ultrasound frequency shift factor of the enhanced analysis target block is obtained.

2. The method according to claim 1, wherein the method is characterized by, After obtaining the ultrasound grayscale image of the wound area to be evaluated, further comprising: Performing noise reduction processing on the ultrasound grayscale image of the wound area to be evaluated to obtain a noise-reduced ultrasound grayscale image.

3. The method according to claim 1, wherein the method is characterized by, The determination of the fat feature screening parameter according to the grayscale distribution uniformity feature and the frequency shift data feature of each enhanced analysis candidate block comprises: For any enhanced analysis candidate block, the grayscale information entropy and the grayscale mean value are determined according to the grayscale value of each pixel point in the enhanced analysis candidate block, and then the first fat feature screening factor is determined; Wherein, the grayscale information entropy and the grayscale mean value are negatively correlated with the first fat feature screening factor; According to the frequency shift data of each pixel point in the enhanced analysis candidate block, the average value of the absolute values of the frequency shift data of all pixel points is calculated to obtain the second fat feature screening factor; Wherein, the average value is negatively correlated with the second fat feature screening factor; and the frequency shift data is the difference between the ultrasound wave transmission frequency and the echo frequency; The first fat feature screening factor and the second fat feature screening factor are combined to determine the fat feature screening parameter of the enhanced analysis candidate block.

4. The method according to claim 1, wherein the method is characterized by, The calculation of the fat echo factor of each enhanced analysis target block comprises: For any enhanced analysis target block, the fitting edge curve of the enhanced analysis target block is obtained, and then the curvature variance of each edge pixel point on the fitting edge curve is calculated; According to the grayscale change feature of the pixel points in the enhanced analysis target block extending to the edge, the effusion ultrasound echo feature value of the enhanced analysis target block is determined; The curvature variance and the effusion ultrasound echo feature value are combined to determine the fat echo factor of the enhanced analysis target block; Wherein, the curvature variance and the effusion ultrasound echo feature value are negatively correlated with the fat echo factor.

5. The method according to claim 4, wherein the method is used for the ultrasonic evaluation of the wound surface of a free flap graft. The determination of the effusion ultrasound echo feature value of the enhanced analysis target block according to the grayscale change feature of the pixel points in the enhanced analysis target block extending to the edge comprises: The geometric centroid coordinate point of the enhanced analysis target block is obtained, and then the connecting line between each pixel point in the enhanced analysis target block and the geometric centroid coordinate point is obtained, and the pixel point in the non-enhanced analysis target block closest to the connecting line direction is taken as the contrast pixel point of the corresponding pixel point in the enhanced analysis target block; The distance value between each pixel point in the enhanced analysis target block and its contrast pixel point is calculated, and the grayscale difference value between each pixel point in the enhanced analysis target block and the geometric centroid coordinate point is calculated; According to the distance value and the gray difference value of each pixel point in the enhanced analysis target block, a liquid ultrasonic echo feature value of the enhanced analysis target block is determined; The distance value is negatively correlated with the liquid ultrasonic echo feature value, and the gray difference value is positively correlated with the liquid ultrasonic echo feature value.

6. The method of claim 1, wherein the method is used for the ultrasonic evaluation of the wound surface in a free flap transplantation. The determination of the fat feature first evaluation value of each enhanced analysis target block comprises: For any enhanced analysis target block, a negative correlation value is obtained by negatively correlating the blood vessel ultrasonic frequency shift feature value of the enhanced analysis target block, and the product of the negative correlation value and the fat echo factor is taken as the fat feature first evaluation value of the enhanced analysis target block.

7. The method of claim 1, wherein the method is used for the ultrasonic evaluation of the wound surface in a free flap transplantation. The determination of the fat feature second evaluation value of each gray level according to the fat feature first evaluation value and the number of pixels corresponding to different gray levels in each enhanced analysis target block comprises: For any enhanced analysis target block, any gray level is taken as a target gray level, and the product of the fat feature first evaluation value and the number of pixel points of the target gray level in the enhanced analysis target block is determined; The accumulated value of the product corresponding to the target gray level in all enhanced analysis target blocks is calculated, the accumulated value is normalized to obtain a normalized value, and the normalized value is taken as the fat feature second evaluation value of the target gray level.

Images