Image enhancement method, device and system for anesthesia puncture

Abstract

The present invention relates to the field of medical image enhancement processing technology, and specifically to an image enhancement method, device and system for anesthesia puncture, which obtains the current ultrasonic image and continuous multiple frames of historical ultrasonic images during the anesthesia puncture process; obtains the shape regularity of each connected domain in the historical ultrasonic image, and combines the grayscale of the pixels in each connected domain to determine the emphasis index of each connected domain, which is used to characterize the possibility of the existence of a puncture needle in the connected domain; obtains the image enhancement degree based on the emphasis index of each connected domain and its stability, and accordingly enhances the current ultrasonic image. The image enhancement degree of each connected domain is obtained according to the characteristics of the historical ultrasonic image, and is reflected in the current ultrasonic image, so as to determine the image enhancement result of each connected domain of the current ultrasonic image, and the enhanced image can improve the imaging clarity of the puncture needle in the image.

Description

Technical Field

[0001] The present invention relates to the technical field of medical image enhancement processing, and in particular to an image enhancement method, device and system for anesthesia puncture. Background Art

[0002] Before using puncture to insert the puncture needle into the lesion for treatment, anesthesia is required to reduce the patient's pain. During the anesthesia puncture process, the doctor uses an ultrasonic probe to obtain an ultrasonic image of the puncture area, and then performs puncture treatment with the help of the ultrasonic image. The puncture needle itself is relatively slender. Therefore, for the ultrasonic image collected in real time during anesthesia, due to the influence of some factors, such as: when holding the ultrasonic probe, the ultrasonic probe will have a certain degree of shaking, thereby affecting the clarity of the puncture needle in the ultrasonic image. In addition, the ultrasonic absorption capacity of the puncture needle is weak, and the ultrasonic absorption capacity of other components or structures is strong, which will also affect the clarity of the puncture needle in the ultrasonic image to a certain extent. If the entire area in the ultrasonic image is directly enhanced, it is possible that other tissues in the ultrasonic image are also enhanced, resulting in the inability to obtain a clear image of the puncture needle in the ultrasonic image. Summary of the invention

[0003] In view of this, in order to solve the problems in the prior art, the present invention provides an image enhancement method, device and system for anesthesia puncture.

[0004] The technical solutions adopted are as follows:

[0005] In a first aspect, an embodiment of the present invention provides an image enhancement method for anesthesia puncture, comprising:

[0006] Acquire a current ultrasonic image during anesthesia puncture, and a plurality of consecutive frames of historical ultrasonic images before and after the current ultrasonic image;

[0007] Obtaining the shape regularity of each connected domain in the historical ultrasonic image, and combining the grayscale of the pixels in each connected domain, determining the emphasis index of each connected domain, wherein the emphasis index is used to characterize the possibility of the presence of a puncture needle in the connected domain;

[0008] Based on the emphasis index and stability of each connected domain for all the historical ultrasonic images, the image enhancement degree of each connected domain is obtained;

[0009] At least based on the image enhancement degree, an image enhancement result of each connected domain in the current ultrasound image is determined.

[0010] In combination with the first aspect above, in some possible implementations, obtaining the shape regularity of each connected domain in the historical ultrasonic image includes:

[0011] Obtaining straight line features of candidate edge pixel points in the connected domain, and determining matching edge pixel points in the edge of the connected domain based on the straight line features; the candidate edge pixel point is any edge pixel point in the connected domain;

[0012] Obtaining a straight line feature difference between the straight line feature of the candidate edge pixel point and the straight line feature of the matching edge pixel point, wherein the straight line feature difference is used to characterize the degree of parallelism between the straight line direction corresponding to the straight line feature of the candidate edge pixel point and the straight line direction corresponding to the straight line feature of the matching edge pixel point;

[0013] The shape irregularity of the connected domain is obtained according to the straight line feature difference of each edge pixel point in the connected domain.

[0014] In combination with the first aspect above, in some possible implementations, the image enhancement method further includes: mapping the current ultrasonic image and each of the historical ultrasonic images into a two-dimensional coordinate system;

[0015] The process of acquiring the straight line features of the candidate edge pixels includes:

[0016] Obtaining a preset number of adjacent edge pixels on both sides of the candidate edge pixel in the edge of the connected domain, and obtaining an angle between a line connecting any two adjacent edge pixels in the candidate edge pixel set and the positive direction of the horizontal axis of the two-dimensional coordinate system; the candidate edge pixel set includes the candidate edge pixel and all the corresponding adjacent edge pixels;

[0017] Calculate the angle average of all angles corresponding to the candidate edge pixel points, where the angle average is the straight line feature of the candidate edge pixel points;

[0018] The process of obtaining the matching edge pixel points includes:

[0019] A straight line corresponding to the average value of the angles of the candidate edge pixel points is drawn through the candidate edge pixel points, and a perpendicular line to the straight line is drawn through the candidate edge pixel points, and the edge pixel point at the intersection of the perpendicular line and the edge of the connected domain is the matching edge pixel point corresponding to the candidate edge pixel point.

[0020] In combination with the first aspect above, in some possible implementations, the grayscale condition of the pixel point is specifically an overall grayscale level of the grayscale values ​​of the pixel points in the connected domain;

[0021] The emphasis index is obtained by the overall grayscale level of the grayscale values ​​of the pixels in the connected domain and the regularity of the shape of the connected domain, wherein the higher the overall grayscale level, the higher the emphasis index; the more regular the shape of the connected domain, the higher the emphasis index.

[0022] In combination with the first aspect, in some possible implementations, based on the emphasis index and stability of each connected domain for all the historical ultrasonic images, the image enhancement degree of each connected domain is obtained, including:

[0023] Obtaining the overall level of the focus index of the candidate connected domain for all the focus indexes of the historical ultrasonic images, and the stability of the focus index, wherein the overall level of the focus index is represented by the average value of the focus index of the candidate connected domain for all the historical ultrasonic images, and the stability is represented by the variance of the focus index of the candidate connected domain for all the historical ultrasonic images and negative correlation; the candidate connected domain is any connected domain;

[0024] The image enhancement degree of the candidate connected domain is obtained by the overall level of the focus index and the stability of the focus index, wherein the higher the overall level of the focus index, the higher the image enhancement degree; the higher the stability of the focus index, the higher the image enhancement degree.

[0025] In combination with the first aspect, in some possible implementations, the image enhancement method further includes: obtaining a position difference of a connected domain in any two adjacent frames of the historical ultrasonic image to obtain an influence index of each connected domain, wherein the influence index is used to characterize the degree of influence of the change in the position of the puncture needle on each connected domain in the current ultrasonic image;

[0026] The determining, at least based on the image enhancement degree, the image enhancement result of each connected domain in the current ultrasonic image includes: determining, based on the image enhancement degree and the influencing index, the image enhancement result of each connected domain in the current ultrasonic image.

[0027] In combination with the first aspect, in some possible implementations, determining the image enhancement result of each connected domain in the current ultrasound image based on the image enhancement degree and the influence index includes:

[0028] According to the image enhancement degree and the influencing index, an image enhancement judgment coefficient of each connected domain in the current ultrasonic image is obtained, wherein the image enhancement judgment coefficient is a normalized value;

[0029] The connected domain corresponding to the image enhancement judgment coefficient greater than or equal to the set threshold is determined as the image enhancement connected domain.

[0030] In combination with the first aspect above, in some possible implementations, the image enhancement method further includes:

[0031] Increasing the image enhancement judgment coefficient of the image enhancement connected domain to obtain an image enhancement amplitude, so that the image enhancement amplitude has a numerical range greater than 1 and less than 2;

[0032] Obtain the product of the initial grayscale value of the candidate pixel point of the image enhancement connected domain and the image enhancement amplitude. If the product is less than 255, use the product as the final grayscale value of the candidate pixel point; if the product is greater than or equal to 255, the final grayscale value of the candidate pixel point is 255; the candidate pixel point is any pixel point in the image enhancement connected domain.

[0033] In a second aspect, an embodiment of the present invention provides an image enhancement device for anesthesia puncture, including an ultrasonic probe and an image processing device, wherein the image processing device is used to execute the above-mentioned image enhancement method for anesthesia puncture.

[0034] In a third aspect, an embodiment of the present invention provides an image enhancement system for anesthesia puncture, comprising a memory and a processor, wherein the memory is used to store executable program code, and the processor is used to call and run the program code from the memory to implement the above-mentioned image enhancement method for anesthesia puncture.

[0035] The present invention has the following technical effects, including but not limited to: selecting multiple frames of continuous historical ultrasonic images before and adjacent to the current ultrasonic image during anesthesia puncture for processing, and enhancing the current ultrasonic image accordingly according to the relevant features shown by the historical ultrasonic images, so as to improve the reliability of image enhancement; since the shape of the puncture needle has obvious regularity relative to other components in the image, the historical ultrasonic images are processed to obtain the shape regularity of each connected domain therein, and since the puncture needle has different absorption capacity for ultrasonic waves relative to other components in the image, which is manifested in different grayscale conditions of pixels in the image, therefore, in combination with various The grayscale of pixels in a connected domain is determined to determine the emphasis index of each connected domain, and the emphasis index is used to characterize the possibility of the existence of a puncture needle in the connected domain; then, the connected domain is taken as the analysis object, and the characteristics presented by the emphasis index of the connected domain in all historical ultrasonic images are analyzed. The characteristics can further accurately characterize the connected domain corresponding to the puncture needle, so as to obtain the image enhancement degree of each connected domain according to the characteristics, and reflect the image enhancement degree of each connected domain in the current ultrasonic image, so as to determine the image enhancement result of each connected domain of the current ultrasonic image, and realize the enhancement of the current ultrasonic image. The enhanced image can improve the imaging clarity of the puncture needle in the image. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a schematic diagram of the hardware composition of an image enhancement device used for anesthesia puncture;

[0037] Figure 2 It is a flow chart of an image enhancement method for anesthesia puncture provided by the present invention;

[0038] Figure 3 It is a flow chart for obtaining the shape regularity of the connected domain;

[0039] Figure 4 It is a flow chart for obtaining the straight line features of candidate edge pixels;

[0040] Figure 5 It is a schematic diagram of a set of candidate edge pixels;

[0041] Figure 6 It is a flowchart for obtaining matching edge pixels;

[0042] Figure 7 is a specific flow chart of step S3;

[0043] Figure 8 It is a structural schematic diagram of an image enhancement system used for anesthesia puncture;

[0044] Fig. 9 It is a schematic diagram of the structure of a computer-readable storage medium;

[0045] Among them, 1 is an ultrasonic probe, 2 is an image processing device, and 3 is a display screen. DETAILED DESCRIPTION

[0046] The device used in the image enhancement method for anesthesia puncture provided in this embodiment is an image enhancement device for anesthesia puncture, such as Figure 1 As shown, the image enhancement device includes an ultrasonic probe 1 and an image processing device 2. As a specific embodiment, the image enhancement device also includes a display screen 3. The ultrasonic probe 1 is signal-connected to the image processing device 2, and the image processing device 2 is signal-connected to the display screen 3. The signal connection in this embodiment can be connected by a wired signal line, or it can be wirelessly connected by a wireless communication module. The ultrasonic probe 1 is used to face the part where anesthesia puncture is required, and to perform imaging using the ultrasonic imaging principle to obtain an ultrasonic image. It should be understood that the ultrasonic probe 1 is aligned with the approximate final position of the puncture needle in the body, and a certain angle is usually formed between the ultrasonic probe 1 and the puncture needle, so that both the target position and the puncture needle can be displayed in the ultrasonic image.

[0047] During the anesthesia puncture process, the ultrasonic probe 1 can acquire continuous multiple frames of ultrasonic images according to a preset sampling period, wherein the sampling period is not fixed and is set according to actual conditions. It should be understood that the time interval between adjacent frames is usually short, such as 0.2 seconds or 0.1 seconds; ultrasonic video can also be acquired, and then continuous multiple frames of ultrasonic images can be acquired through image frame extraction based on the ultrasonic video.

[0048] The ultrasonic probe 1 sends the collected continuous multi-frame ultrasonic images to the image processing device 2, which is used to perform image processing. The hardware device corresponding to the image processing device 2 can be a computer host or server device with certain image processing capabilities. The image processing device 2 sends the processed images to the display screen 3, which displays them for relevant personnel to view. The image processing device 2 is used to perform Figure 2 An image enhancement method for anesthesia puncture (hereinafter referred to as image enhancement method) is shown.

[0049] like Figure 2 As shown, the image enhancement method includes the following steps:

[0050] Step S1: Acquire a current ultrasonic image during anesthesia puncture, and a plurality of consecutive frames of historical ultrasonic images immediately before the current ultrasonic image.

[0051] During the anesthesia puncture process, the ultrasonic image that currently needs image enhancement is used as the current ultrasonic image, and multiple consecutive frames of historical ultrasonic images before the current ultrasonic image are obtained. The number of frames of the historical ultrasonic images obtained is set according to actual needs, such as 30 frames. Moreover, the last frame of the historical ultrasonic image in the continuous multiple frames of historical ultrasonic images obtained is the previous frame of the current ultrasonic image, so that the continuous multiple frames of historical ultrasonic images and the current ultrasonic image are continuous in time sequence, thereby obtaining the continuous multiple frames of historical ultrasonic images before the current ultrasonic image.

[0052] In this embodiment, each ultrasonic image is a grayscale image. If the initially acquired ultrasonic image is not a grayscale image, it needs to be converted into a grayscale image.

[0053] Step S2: Obtain the shape regularity of each connected domain in the historical ultrasonic image, and determine the emphasis index of each connected domain in combination with the grayscale of the pixels in each connected domain, wherein the emphasis index is used to characterize the possibility of the presence of a puncture needle in the connected domain.

[0054] Connected domain processing is performed on each ultrasonic image, including the current ultrasonic image and each historical ultrasonic image, to obtain each connected domain in each ultrasonic image. It should be understood that the present embodiment can use the existing connected domain extraction algorithm to obtain the connected domain in the image, for example: based on the existing dbscan clustering method, the neighborhood radius is set to 5 according to the empirical value, and the minimum number of samples is set to 5, and all the collected ultrasonic images are processed to obtain each connected domain in each ultrasonic image, or the Otsu threshold algorithm is used to segment each ultrasonic image to obtain each connected domain in each ultrasonic image.

[0055] The shape of the puncture needle is usually a rectangle or a conical structure close to a rectangle. In the ultrasound image, the shape of the puncture needle is more regular than other components or tissues in the image. As for other tissues or structures in the ultrasound image, due to the influence of relatively complex spatial shapes, the shapes of other tissues in the ultrasound image are somewhat different from the shape of the puncture needle. Therefore, for any historical ultrasound image, the shape regularity of each connected domain in the historical ultrasound image is obtained.

[0056] Moreover, it is known a priori that the puncture needle has a weak absorption capacity for ultrasound, so the reflected signal is strong and its grayscale value in the ultrasound image is high. Therefore, the grayscale of the pixels in each connected domain is obtained.

[0057] Therefore, according to the shape regularity of each connected domain in the historical ultrasonic image and the grayscale of the pixels in the connected domain, the possibility of the existence of the puncture needle in each connected domain is determined and characterized by the emphasis index. The higher the emphasis index, the higher the possibility of the existence of the puncture needle in the connected domain.

[0058] Step S3: based on the emphasis index and stability of each connected domain for all the historical ultrasonic images, the image enhancement degree of each connected domain is obtained.

[0059] During the anesthesia puncture process, as the angle of the puncture needle is adjusted, the position of the puncture needle in the ultrasonic image gradually changes. The shape of the human body structure or tissue will also change with the angle of the puncture probe. However, the changes in the human body structure or tissue are relatively random changes, such as the bones and muscles in the spine. However, the puncture needle will only change more regularly as described in the above logic, and has a cone or rectangular shape as a whole. Moreover, the time interval between adjacent historical ultrasonic images is usually short, and the shape of the puncture needle changes less. Therefore, the emphasis index of the connected domain for the puncture needle based on the above calculation is always at a high level during the process of image frame number changes, and the value fluctuates less. Therefore, according to the emphasis index of each connected domain for all historical ultrasonic images and its stability, the image enhancement degree of each connected domain for the current ultrasonic image can be obtained.

[0060] Step S4: determining the image enhancement results of each connected domain in the current ultrasonic image at least based on the image enhancement degree.

[0061] According to the obtained image enhancement degree of each connected domain for the current ultrasonic image, each connected domain of the current ultrasonic image is enhanced according to the corresponding image enhancement degree, and the image enhancement result of each connected domain in the current ultrasonic image is obtained, thereby achieving the enhancement of the current ultrasonic image. As a specific implementation, a threshold can be set to compare the size relationship between the image enhancement degree of each connected domain and the threshold, and the image area of ​​the connected domain corresponding to the image enhancement degree greater than the threshold in the current ultrasonic image is enhanced. It should be understood that the algorithm for image enhancement of the image area can adopt an existing conventional image enhancement algorithm, such as a grayscale histogram equalization algorithm.

[0062] In a specific embodiment, Figure 3 As shown in the figure, the process of obtaining the shape regularity of the connected domain is given as follows:

[0063] Step S2-1: Obtain straight line features of candidate edge pixel points in the connected domain, and determine matching edge pixel points in the edge of the connected domain based on the straight line features; the candidate edge pixel point is any edge pixel point in the connected domain.

[0064] Since the edge of a connected domain is composed of multiple edge pixels, for ease of description, the candidate edge pixel is set to be any edge pixel in the connected domain to process all edge pixels in the connected domain.

[0065] The straight line features of the candidate edge pixels are obtained. The straight line features are used to characterize the relevant features of the straight line at the location of the candidate edge pixels. The straight line features of the candidate edge pixels are used to obtain matching edge pixels that match the candidate edge pixels, and the shape regularity of the connected domain is determined by combining the straight line features of the matching edge pixels.

[0066] In this embodiment, the horizontal and vertical directions of the ultrasonic image are used as the horizontal axis and vertical axis of the two-dimensional coordinate system, respectively, to construct a two-dimensional coordinate system. The current ultrasonic image and each historical ultrasonic image are mapped to the two-dimensional coordinate system, and then the position of each pixel point of each ultrasonic image in the image is the two-dimensional coordinate point in the two-dimensional coordinate system.

[0067] like Figure 4 As shown in FIG. 1 , a specific process of obtaining the straight line features of the candidate edge pixels is given, including:

[0068] Step S2-1-1: Obtain a preset number of adjacent edge pixel points on both sides of the candidate edge pixel point in the edge of the connected domain, and obtain the angle between the line connecting any two edge pixel points in the candidate edge pixel point set and the positive direction of the horizontal axis of the two-dimensional coordinate system; the candidate edge pixel point set includes the candidate edge pixel point and all the corresponding adjacent edge pixel points.

[0069] Among them, the candidate edge pixel set has a corresponding relationship with the candidate edge pixel point, and the candidate edge pixel set of the candidate edge pixel point includes the candidate edge pixel point and all adjacent edge pixel points corresponding to the candidate edge pixel point. The preset number is set according to actual needs. For example, in the edge of the connected domain, the 5 edge pixel points adjacent to both sides of the candidate edge pixel point are called adjacent edge pixel points, that is, the 5 adjacent edge pixel points adjacent to the left side of the candidate edge pixel point, and the 5 adjacent edge pixel points adjacent to the right side. Then, the candidate edge pixel set of the candidate edge pixel point includes the candidate edge pixel point and 10 adjacent edge pixel points corresponding to the candidate edge pixel point, such as Figure 5 As shown, Figure 5 Each square in represents an edge pixel point, wherein the square with a shadow represents a candidate edge pixel point, and the five adjacent edge pixels on both sides of the candidate edge pixel point are respectively the five adjacent edge pixels on the left side of the candidate edge pixel point, and the five adjacent edge pixels on the right side, which together constitute the candidate edge pixel point set of the candidate edge pixel point.

[0070] Then, any two adjacent edge pixel points in the candidate edge pixel point set are connected to form a straight line, thereby obtaining the connecting lines of any two adjacent edge pixel points in the candidate edge pixel point set. Each connecting line will intersect with the positive direction of the horizontal axis of the two-dimensional coordinate system, thereby generating an angle, thereby obtaining multiple angles corresponding to the candidate edge pixel point set, that is, obtaining multiple angles corresponding to the candidate edge pixel point. It should be understood that if the connecting line is exactly parallel to the horizontal axis of the two-dimensional coordinate system, that is, it does not intersect, then the angle is 0.

[0071] Step S2-1-2: Calculate the angle average of all angles corresponding to the candidate edge pixel points, where the angle average is the straight line feature of the candidate edge pixel points.

[0072] By adopting the above process, the straight line features of each edge pixel point in the edge of the connected domain are obtained.

[0073] A matching edge pixel point that matches the candidate edge pixel point is obtained from the edge. In this embodiment, the matching edge pixel point is obtained as follows: Figure 6 As shown, including:

[0074] Step S2-1-3: Draw a straight line through the candidate edge pixel point that corresponds to the average value of the angles of the candidate edge pixel point.

[0075] Among them, since the angle average value represents the angle between the positive direction of the horizontal axis of the two-dimensional coordinate system, after obtaining the angle average value of the candidate edge pixel point, the direction of the angle average value is also obtained, and a straight line can be drawn through the candidate edge pixel point in the same direction as the angle average value of the candidate edge pixel point.

[0076] Step S2-1-4: Draw a perpendicular line to the straight line through the candidate edge pixel point, and the edge pixel point at the intersection of the perpendicular line and the edge of the connected domain is the matching edge pixel point corresponding to the candidate edge pixel point.

[0077] The obtained vertical line will intersect with the edge of the connected domain, and in addition to the candidate edge pixel point, another edge pixel point at the intersection can be obtained, and the edge pixel point at the other intersection is used as the matching edge pixel point corresponding to the candidate edge pixel point. Then, the straight line feature of the matching edge pixel point is obtained.

[0078] Step S2-2: Obtain the straight line feature difference between the straight line feature of the candidate edge pixel point and the straight line feature of the matching edge pixel point, wherein the straight line feature difference is used to characterize the degree of parallelism between the straight line direction corresponding to the straight line feature of the candidate edge pixel point and the straight line direction corresponding to the straight line feature of the matching edge pixel point.

[0079] The straight line feature of the edge pixel essentially represents the direction of the straight line where the edge pixel is located. Then, the straight line feature difference between the straight line feature of the candidate edge pixel and the straight line feature of the matching edge pixel represents the difference between the directions of the straight lines where the two edge pixels are located. The smaller the difference, the closer the directions are and the more parallel the two straight lines are. Conversely, the greater the difference in directions, the less parallel the two straight lines are. Then, the straight line feature difference between the straight line feature of the candidate edge pixel and the straight line feature of the matching edge pixel is obtained. The straight line feature difference represents the degree of parallelism between the straight line direction corresponding to the straight line feature of the candidate edge pixel and the straight line direction corresponding to the straight line feature of the matching edge pixel. The straight line feature difference between the straight line feature of the candidate edge pixel and the straight line feature of the matching edge pixel is taken as the straight line feature difference of the candidate edge pixel. Thus, the straight line feature difference of each edge pixel is obtained.

[0080] In this embodiment, the straight line feature difference is specifically the absolute value of the difference between the straight line feature of the candidate edge pixel point and the straight line feature of the matching edge pixel point.

[0081] Step S2-3: Obtaining the shape irregularity of the connected domain according to the straight line feature difference of each edge pixel point in the connected domain.

[0082] The average value of the straight line feature differences of all edge pixels in the connected domain is calculated as the shape irregularity of the connected domain. The larger the average value of the straight line feature differences, the higher the shape irregularity of the connected domain, that is, the more irregular the shape of the connected domain.

[0083] In a specific implementation, the grayscale condition of the pixel points in step S2 is specifically the overall grayscale level of the grayscale values ​​of the pixel points in the connected domain, and the overall grayscale level can be represented as the average value of the grayscale values ​​of all the pixel points in the connected domain.

[0084] Then, the emphasis index of the connected domain is obtained by the overall grayscale level of the grayscale values ​​of the pixels in the connected domain and the shape regularity of the connected domain, wherein the higher the overall grayscale level, the higher the emphasis index; the more regular the shape of the connected domain, the higher the emphasis index. Since the shape regularity is specifically manifested as shape irregularity, then the emphasis index is obtained by the overall grayscale level of the grayscale values ​​of the pixels in the connected domain and the shape irregularity of the connected domain, wherein the higher the overall grayscale level, the higher the emphasis index; the lower the shape irregularity of the connected domain, the higher the emphasis index. Based on the above logic, this embodiment characterizes the above logical relationship in the form of parameter subtraction, specifically: normalize the overall grayscale level and shape irregularity respectively, then obtain the difference between the normalized overall grayscale level and the normalized shape irregularity, add the obtained difference to 1 and divide by 2, and the result is used as the emphasis index of the connected domain. Since the difference between two values ​​in [0,1] is in the range of [-1,1], then the difference is added to 1 and divided by 2, and the value range of the result is [0,1]. It should be understood that the normalization method in this embodiment can be a maximum and minimum value normalization method, or it can be the following normalization method: , where y is the normalized value, x is the input value, and exp is the exponential function with the natural constant e as the base.

[0085] By adopting the above process, the emphasis index of each connected domain in each historical ultrasonic image is obtained.

[0086] In a specific embodiment, Figure 7 A specific implementation process of step S3 is given, including:

[0087] Step S3-1: Obtain the overall level of the focus indicators of the candidate connected domain for all the historical ultrasonic images, as well as the stability of the focus indicators, wherein the overall level of the focus indicators is represented by the average value of the focus indicators of the candidate connected domain for all the historical ultrasonic images, and the stability is represented by the variance of the focus indicators of the candidate connected domain for all the historical ultrasonic images and negative correlation; the candidate connected domain is any connected domain.

[0088] For ease of explanation, the candidate connected domain is set to any connected domain. It should be understood that during the puncture anesthesia process, due to the time interval between each frame of ultrasonic image (including each historical ultrasonic image and the current ultrasonic image), the position of some connected domains in the ultrasonic image may change, such as the position of the puncture needle. Because the time interval between adjacent frames of ultrasonic images is relatively short, even if the position of the connected domain changes, the change amplitude is usually not too large. Therefore, for the candidate connected domain in the previous frame of ultrasonic image in any two adjacent frames of ultrasonic images, the distance between the centroid of each connected domain in the next frame of ultrasonic image and the centroid of the candidate connected domain is obtained, and the minimum distance is obtained therefrom, and the connected domain corresponding to the minimum distance in the next frame of ultrasonic image is used as the matching connected domain of the candidate connected domain, thereby obtaining the matching connected domain of the candidate connected domain in each ultrasonic image according to the time sequence. Each matching connected domain is determined to be the same connected domain as the candidate connected domain, thereby obtaining the candidate connected domains in all ultrasonic images.

[0089] Therefore, each candidate connected domain has a corresponding emphasis index in each historical ultrasonic image, and then, for all historical ultrasonic images, the candidate connected domain will obtain an emphasis index set corresponding to the candidate connected domain, and the emphasis index set includes the emphasis index of the candidate connected domain in each historical ultrasonic image. More preferably, the emphasis indexes in the emphasis index set can be sorted according to the image frame order of the historical ultrasonic image, thereby obtaining a emphasis index sequence.

[0090] The overall level of emphasis indicators of the candidate connected domain for all historical ultrasonic images is obtained, that is, the overall level of emphasis indicators of the emphasis indicator set corresponding to the candidate connected domain is obtained, wherein the overall level of emphasis indicators is represented by the average value of the emphasis indicators of the candidate connected domain for all historical ultrasonic images, that is, the average value of the emphasis indicators in the emphasis indicator set is calculated, and the average value is used as the overall level of emphasis indicators.

[0091] Obtain the stability of the focus index set of the candidate connected domain. Since the variance represents the fluctuation of the data set, in this embodiment, the variance of the focus index set of the candidate connected domain is first obtained, and then the variance is negatively correlated to obtain the stability of the focus index.

[0092] Step S3-2: The image enhancement degree of the candidate connected domain is obtained by the overall level of the focus index and the stability of the focus index, wherein the higher the overall level of the focus index, the higher the image enhancement degree; the more stable the focus index, that is, the higher the stability of the focus index, the higher the image enhancement degree.

[0093] This embodiment provides a specific quantification method: since the numerical range of the variance of the focus indicator set of the candidate connected domain is [0, 0.25], the variance is multiplied by 4, its numerical range is expanded to [0, 1], and then the expanded variance is subtracted from the value 1, and the result is used as the stability of the focus indicator. This processing method not only realizes the negative correlation of the variance, but also stabilizes the numerical range of the focus indicator stability at [0, 1]. Finally, the product of the overall level of the focus indicator and the stability of the focus indicator is calculated, and the obtained product is the image enhancement degree of the candidate connected domain. Through the above method, the image enhancement degree can be stabilized in [0, 1], which is convenient for subsequent calculations and comparisons with thresholds.

[0094] In addition, according to the analysis, during the puncture anesthesia process, the puncture needle will move to a certain extent, and among the components or tissues other than the puncture needle, the tissue in contact with the puncture needle is affected by the movement of the puncture needle, and the connected domain will change slightly, while the connected domain of the tissue far from the puncture needle will hardly change. Therefore, the position change of the connected domain can also be combined to characterize the degree to which each connected domain in the current ultrasound image is affected by the change in the position of the puncture needle.

[0095] Therefore, in a specific embodiment, the image enhancement method further includes: obtaining the position difference of the connected domain in any two adjacent historical ultrasonic images, and obtaining the influence index of each connected domain, and the influence index is used to characterize the degree of influence of the change in the position of the puncture needle on each connected domain in the current ultrasonic image. Specifically: for the candidate connected domain, the position difference of the candidate connected domain in each two adjacent ultrasonic images is obtained. In this embodiment, the candidate connected domain in the two adjacent ultrasonic images has two centroids, then the position difference is the distance between the two centroids of the candidate connected domain in the two adjacent ultrasonic images in the two-dimensional coordinate system. Thus, multiple position differences of the candidate connected domain are obtained to form a position difference sequence, and then the average value of the position difference in the position difference sequence of the candidate connected domain is calculated, and then the average value of the position difference is normalized, and the normalized result is used as the influence index of the candidate connected domain. Using the above process, the influence index of each connected domain is obtained.

[0096] The larger the impact index is, the greater the degree of position change of the candidate connected domain during the puncture anesthesia process, the greater the degree of influence of the change in the position of the puncture needle, the more important the image area in the ultrasound image is, and the more image enhancement is needed.

[0097] Correspondingly, step S4 specifically includes: determining the image enhancement result of each connected domain in the current ultrasonic image based on the image enhancement degree and the influence index.

[0098] As a specific implementation, the image enhancement judgment coefficient of each connected domain in the current ultrasonic image is obtained according to the image enhancement degree and the influence index. Among them, for the candidate connected domain in the current ultrasonic image, the image enhancement judgment coefficient of the candidate connected domain is obtained according to the image enhancement degree of the candidate connected domain and the influence index of the candidate connected domain. Since the image enhancement degree and the influence index are both positively correlated with the image enhancement judgment coefficient, this embodiment uses the product of the image enhancement degree of the candidate connected domain and the influence index of the candidate connected domain as the image enhancement judgment coefficient of the candidate connected domain. Since the image enhancement degree and the influence index are both normalized values, the obtained image enhancement judgment coefficient can also be understood as a normalized value. The larger the image enhancement judgment coefficient, the more image enhancement is needed.

[0099] A threshold is set, and the value range of the threshold is [0, 1]. The specific value is set according to the actual image enhancement needs. For example, the stricter the image enhancement requirements, the larger the value of the threshold is. This embodiment takes 0.7 as an example.

[0100] The image enhancement judgment coefficients of each connected domain are compared with the set threshold, and the image enhancement judgment coefficients greater than or equal to the set threshold are obtained, thereby obtaining the connected domains of these image enhancement judgment coefficients, and these connected domains of the current ultrasonic image are used as the image enhancement connected domains in the current ultrasonic image. The image enhancement connected domain in the current ultrasonic image is a possible puncture needle area.

[0101] The image enhancement of the image enhancement connected domain can be performed by using an existing commonly used image enhancement method such as a histogram equalization algorithm. In this embodiment, the purpose of image enhancement of the image enhancement connected domain is to improve its contrast with other surrounding image regions. In this embodiment, the grayscale value of the pixel is appropriately increased to perform image enhancement. Then, the image enhancement method also includes the following image enhancement process: for any image enhancement connected domain, the image enhancement judgment coefficient of the image enhancement connected domain is numerically increased. The purpose of the numerical increase is to make the numerical range of the image enhancement amplitude greater than 1 and less than 2. As a specific implementation method, the image enhancement judgment coefficient is added to the value 1 to increase it to greater than 1 and less than 2. The result of the numerical increase is the image enhancement amplitude. Set the candidate pixel point as any pixel point in the image enhancement connected domain, obtain the product of the initial grayscale value of the candidate pixel point of the image enhancement connected domain and the image enhancement amplitude of the image enhancement connected domain, if the product is less than 255, then the product is used as the final grayscale value of the candidate pixel point; if the product is greater than or equal to 255, then the final grayscale value of the candidate pixel point is 255. The final grayscale value of the candidate pixel replaces the initial grayscale value of the candidate pixel to achieve image enhancement of the image enhancement connected domain. The above method is used to achieve image enhancement of each image enhancement connected domain and complete image enhancement of the current ultrasonic image.

[0102] This embodiment also provides an image enhancement system 700 for anesthesia puncture. Figure 8 As shown, it includes a memory 701 and a processor 702, the memory 701 is used to store an executable program code 703, and the processor 702 is used to call and run the program code 703 from the memory 701 to implement Figure 2 The image enhancement method for anesthesia puncture is shown.

[0103] This embodiment also provides a computer-readable storage medium 800, such as Fig. 9 As shown, the computer readable storage medium 800 stores a computer program code 801. When the computer program code 801 is executed on a computer, the computer executes Figure 2 The image enhancement method for anesthesia puncture is shown.

[0104] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. An image enhancement method for anesthesia puncture, characterized in that: include: Acquire a current ultrasonic image during anesthesia puncture, and a plurality of consecutive frames of historical ultrasonic images before and after the current ultrasonic image; Obtaining the shape regularity of each connected domain in the historical ultrasonic image, and combining the grayscale of the pixels in each connected domain, determining the emphasis index of each connected domain, wherein the emphasis index is used to characterize the possibility of the existence of the puncture needle in the connected domain, the shape regularity is used to characterize the difference between the shape of other tissues in the ultrasonic image and the shape of the puncture needle, and the grayscale of the pixel is specifically the overall grayscale level of the grayscale value of the pixel in the connected domain; Based on the emphasis index and the stability of the emphasis index of each connected domain for all the historical ultrasonic images, the image enhancement degree of each connected domain is obtained, and the stability of the emphasis index is characterized by the variance of the emphasis index of the candidate connected domain for all the historical ultrasonic images and negative correlation, and the candidate connected domain is any connected domain; Based on the image enhancement degree, an image enhancement result of each connected domain in the current ultrasonic image is determined.

2. The image enhancement method for anesthesia puncture according to claim 1, characterized in that: Obtaining shape regularity of each connected domain in the historical ultrasonic image includes: Obtaining straight line features of candidate edge pixel points in the connected domain, and determining matching edge pixel points in the edge of the connected domain based on the straight line features; the candidate edge pixel point is any edge pixel point in the connected domain; Obtaining a straight line feature difference between the straight line feature of the candidate edge pixel point and the straight line feature of the matching edge pixel point, wherein the straight line feature difference is used to characterize the degree of parallelism between the straight line direction corresponding to the straight line feature of the candidate edge pixel point and the straight line direction corresponding to the straight line feature of the matching edge pixel point; The shape irregularity of the connected domain is obtained according to the straight line feature difference of each edge pixel point in the connected domain.

3. The image enhancement method for anesthesia puncture according to claim 2, characterized in that: The image enhancement method further includes: mapping the current ultrasonic image and each of the historical ultrasonic images into a two-dimensional coordinate system; The process of acquiring the straight line features of the candidate edge pixels includes: Obtaining a preset number of adjacent edge pixels on both sides of the candidate edge pixel in the edge of the connected domain, and obtaining an angle between a line connecting any two adjacent edge pixels in the candidate edge pixel set and the positive direction of the horizontal axis of the two-dimensional coordinate system; the candidate edge pixel set includes the candidate edge pixel and all the corresponding adjacent edge pixels; Calculate the angle average of all angles corresponding to the candidate edge pixel points, where the angle average is the straight line feature of the candidate edge pixel points; The process of obtaining the matching edge pixel points includes: The angle average value represents the angle with the positive direction of the horizontal axis of the two-dimensional coordinate system. According to the angle average value of the candidate edge pixel points, the direction of the angle average value is obtained, and a straight line consistent with the direction of the angle average value of the candidate edge pixel points is drawn through the candidate edge pixel points, and a perpendicular line to the straight line is drawn through the candidate edge pixel points. The edge pixel point at the intersection of the perpendicular line and the edge of the connected domain is the matching edge pixel point corresponding to the candidate edge pixel point.

4. The image enhancement method for anesthesia puncture according to claim 1, characterized in that: The emphasis index is obtained by the overall grayscale level of the grayscale values ​​of the pixels in the connected domain and the regularity of the shape of the connected domain, wherein the higher the overall grayscale level, the higher the emphasis index; the more regular the shape of the connected domain, the higher the emphasis index; The step of obtaining the emphasis index is: normalizing the overall gray level and shape irregularity respectively, then obtaining the difference between the normalized overall gray level and the normalized shape irregularity, adding the obtained difference to 1 and dividing by 2, and obtaining the result as the emphasis index of the connected domain.

5. The image enhancement method for anesthesia puncture according to claim 1, characterized in that: Based on the emphasis index and the stability of the emphasis index of each connected domain for all the historical ultrasonic images, the image enhancement degree of each connected domain is obtained, including: Obtaining the overall level of the focus index of the candidate connected domain for all the focus indexes of the historical ultrasonic images, and the stability of the focus index, wherein the overall level of the focus index is represented by the average value of the focus index of the candidate connected domain for all the historical ultrasonic images; The image enhancement degree of the candidate connected domain is obtained by the overall level of the focus index and the stability of the focus index, wherein the higher the overall level of the focus index, the higher the image enhancement degree; the higher the stability of the focus index, the higher the image enhancement degree.

6. The image enhancement method for anesthesia puncture according to claim 1, characterized in that: The image enhancement method further includes: obtaining the position difference of the connected domains in any two adjacent frames of the historical ultrasonic images to obtain an influence index of each connected domain, wherein the influence index is used to characterize the degree of influence of the puncture needle position change on each connected domain in the current ultrasonic image; Determining the image enhancement results of each connected domain in the current ultrasonic image based on the image enhancement degree includes: determining the image enhancement results of each connected domain in the current ultrasonic image based on the image enhancement degree and the influence index; The step of obtaining the position difference of the connected domains in any two adjacent frames of the historical ultrasonic images and obtaining the influence index of each connected domain comprises: For the candidate connected domain, the position difference of the candidate connected domain in each two adjacent ultrasonic image frames is obtained. The candidate connected domain in two adjacent ultrasonic image frames has two centroids, and the position difference is the distance between the two centroids of the candidate connected domain in the two adjacent ultrasonic image frames in the two-dimensional coordinate system, thereby obtaining multiple position differences of the candidate connected domain to form a position difference sequence; the average value of the position differences in the position difference sequence of the candidate connected domain is calculated, and then the average value of the position differences is normalized, and the normalized result is used as the influence indicator of the candidate connected domain.

7. The image enhancement method for anesthesia puncture according to claim 6, characterized in that: Based on the image enhancement degree and the influencing index, determining the image enhancement result of each connected domain in the current ultrasonic image includes: According to the image enhancement degree and the influencing index, an image enhancement judgment coefficient of each connected domain in the current ultrasonic image is obtained, wherein the image enhancement judgment coefficient is a normalized value; Determine a connected domain corresponding to an image enhancement judgment coefficient that is greater than or equal to a set threshold as an image enhancement connected domain; The step of obtaining the image enhancement judgment coefficient of each connected domain in the current ultrasonic image according to the image enhancement degree and the influence index includes: taking the product of the image enhancement degree of the candidate connected domain and the influence index of the candidate connected domain as the image enhancement judgment coefficient of the candidate connected domain.

8. The image enhancement method for anesthesia puncture according to claim 7, characterized in that: The image enhancement method further comprises: Increasing the image enhancement judgment coefficient of the image enhancement connected domain to obtain an image enhancement amplitude, so that the image enhancement amplitude has a numerical range greater than 1 and less than 2; Obtain the product of the initial grayscale value of the candidate pixel point of the image enhancement connected domain and the image enhancement amplitude. If the product is less than 255, use the product as the final grayscale value of the candidate pixel point; if the product is greater than or equal to 255, the final grayscale value of the candidate pixel point is 255; the candidate pixel point is any pixel point in the image enhancement connected domain.

9. An image enhancement device for anesthesia puncture, comprising an ultrasonic probe and an image processing device, characterized in that: The image processing device is used to execute the image enhancement method for anesthesia puncture according to any one of claims 1-8.

10. An image enhancement system for anesthesia puncture, characterized in that: It comprises a memory and a processor, wherein the memory is used to store executable program code, and the processor is used to call and run the program code from the memory to implement the image enhancement method for anesthesia puncture as described in any one of claims 1-8.

Images