A method and system for clinical anesthesia ultrasound image-assisted positioning guidance

Abstract

The present invention relates to the field of image enhancement technology, and specifically to a method and system for assisting positioning and guiding with ultrasound images for clinical anesthesia, comprising: obtaining a needle centerline and an enhanced demarcation distance in a lumbar ultrasound image; obtaining main edge pixel points according to the lumbar ultrasound image; obtaining the probability of a lamina edge according to the grayscale values ​​of the main edge pixel points and the pixel points; obtaining a suspected lamina area according to the correlation value and the lamina edge probability; obtaining a relative position angle and a lamina-related angle according to the suspected lamina area; obtaining a grayscale enhancement coefficient according to the relative position angle, the lamina-related angle, and the extension trend; obtaining an ultrasound-enhanced image and performing assisting positioning and guiding according to the grayscale enhancement coefficient, the needle centerline, and the enhanced demarcation distance. The present invention improves the assisting positioning and guiding effect of ultrasound-guided spinal anesthesia by enhancing the needle area in the lumbar ultrasound image and reducing the impact on the lamina.

Description

Technical Field

[0001] The present invention relates to the technical field of image enhancement, and in particular to a method and system for clinical anesthesia ultrasonic image-assisted positioning guidance. Background Art

[0002] Spinal anesthesia is a commonly used anesthesia method in clinical practice. It has the advantages of rapid onset and definite effect. It is often used in lower limb and abdominal and pelvic surgery. Accurate positioning of the puncture point is the first step in implementing spinal anesthesia. The traditional positioning method is surface palpation positioning, but due to individual anatomical variations, it often leads to difficulties in manual positioning, puncture failure, and even serious complications. Ultrasound examination can intuitively display the anatomical structure of the spinal canal and its surrounding tissues, measure the distance from the skin to the spinal canal, and improve the success rate of puncture. Because the anatomical landmarks of obese, elderly, scoliotic, and previous spinal surgery patients are difficult to palpate, lumbar ultrasound images are used to assist in positioning the patient's puncture point.

[0003] During ultrasound-guided spinal anesthesia, the needle puncture position is adjusted mainly based on the relative position relationship between the needle and the anatomical landmark in the lumbar ultrasound image, where the anatomical landmark is mainly the lamina area. Due to the different types and densities of soft tissues, the grayscale value distribution of the soft tissue area in the lumbar ultrasound image is uneven, resulting in unclear needle imaging. When the grayscale of the needle area is enhanced and the grayscale of the area adjacent to the needle is suppressed, if there is a lamina area in the area adjacent to the needle, it will affect the positioning of the needle and reduce the accuracy of the puncture point positioning. Summary of the invention

[0004] The invention provides a clinical anesthesia ultrasound image-assisted positioning and guiding method and system to solve the existing problems.

[0005] A clinical anesthesia ultrasound image-assisted positioning guidance method and system of the present invention adopts the following technical solutions:

[0006] The present invention proposes a clinical anesthesia ultrasound image-assisted positioning and guiding method, which comprises the following steps:

[0007] The needle centerline and enhancement demarcation distance in lumbar ultrasound images were obtained;

[0008] According to the grayscale distribution of the pixels in the lumbar ultrasound image, the main edge pixels are obtained; according to the grayscale values ​​of the main edge pixels and the pixels in the lumbar ultrasound image, the lamina edge probability of the main edge pixels is obtained;

[0009] According to the correlation values ​​of the pixels in the lumbar ultrasound image with all the main edge pixels and the lamina edge probabilities of all the main edge pixels, the pixels are classified to obtain the suspected lamina area;

[0010] According to the morphological similarity of the suspected lamina area, the extension trend of the suspected lamina area is obtained; according to the relative position distribution of the suspected lamina area, the relative position angle of the suspected lamina area is obtained, and according to the relative position angle and extension trend of the suspected lamina area, the lamina-related angle is obtained; according to the similarity between the relative position angle of the suspected lamina area and the lamina-related angle, combined with the extension trend of the suspected lamina area, the grayscale enhancement coefficient of the pixel point is obtained;

[0011] According to the grayscale enhancement coefficient of the pixel point, the needle centerline and the enhancement boundary distance, an ultrasonic enhanced image is obtained and auxiliary positioning guidance is performed.

[0012] Furthermore, the method of obtaining the main edge pixel points according to the grayscale distribution of the pixel points in the lumbar ultrasound image includes the following specific steps:

[0013] The lumbar ultrasound image is threshold segmented and the binary image hole filling is performed to obtain the lumbar binary image; in the two grayscale areas of the lumbar binary image, the area with the largest average grayscale value is recorded as the high echo area; the edge pixel points of the high echo area are recorded as the main edge pixel points.

[0014] Furthermore, the lamina edge probability of the main edge pixel point is obtained according to the gray value of the main edge pixel point and the pixel point in the lumbar ultrasound image, and the specific steps include the following:

[0015] Any pixel point in the lumbar ultrasound image is recorded as a target pixel point; for any main edge pixel point, the inverse normalization result of the grayscale variance of all pixels on the straight line path from the target pixel point to the main edge pixel point is recorded as the correlation value of the target pixel point to the main edge pixel point;

[0016] The product of the normalized result of the correlation value of the target pixel point with respect to the main edge pixel point and the gray value of the target pixel point is recorded as the probability factor of the target pixel point with respect to the main edge pixel point; the normalized result of the sum of the probability factors of all pixels in the lumbar ultrasound image with respect to the main edge pixel point is recorded as the lamina edge probability of the main edge pixel point.

[0017] Furthermore, the pixel points are classified according to the correlation values ​​of the pixel points in the lumbar ultrasound image with respect to all the main edge pixel points and the lamina edge probabilities of all the main edge pixel points to obtain the suspected lamina area, and the specific steps include the following:

[0018] Any pixel in the lumbar ultrasound image is recorded as a target pixel; for any main edge pixel, the product of the lamina edge probability of the main edge pixel and the correlation value of the target pixel to the main edge pixel is recorded as the possible factor of the target pixel to the main edge pixel; the sum of the possible factors of the target pixel to all the main edge pixels is recorded as the lamina highlight possibility of the target pixel;

[0019] The pixel points in the lumbar ultrasound image are divided into two categories according to the possibility of lamina highlighting using the threshold segmentation method. The pixel points in the category with the greatest possibility of lamina highlighting are recorded as suspected lamina pixels; the area composed of suspected lamina pixels in the lumbar ultrasound image is recorded as suspected lamina area.

[0020] Furthermore, the step of obtaining the extension trend of the suspected lamina region according to the morphological similarity of the suspected lamina region includes the following specific steps:

[0021] Extracting skeleton lines of each suspected lamina region using a skeleton extraction algorithm; obtaining a main extension angle of the suspected lamina region according to the direction of the skeleton lines;

[0022] Any suspected lamina area is recorded as the target suspected lamina area, and the cosine value of the angle corresponding to the absolute value of the difference between the main extension angles of the target suspected lamina area and any other suspected lamina area is recorded as the trend factor between the target suspected lamina area and the suspected lamina area; the average of the trend factors of the target suspected lamina area and all suspected lamina areas is recorded as the extension trend of the target suspected lamina area.

[0023] Further, the relative position angle of the suspected lamina area is obtained according to the relative position distribution of the suspected lamina area, and the lamina-related angle is obtained according to the relative position angle and extension trend of the suspected lamina area, including the following specific steps:

[0024] Obtain the skeleton line and the midpoint of each suspected lamina region; obtain the relative position angle of any two suspected lamina regions according to the midpoints of the skeleton lines of the two suspected lamina regions;

[0025] Any two suspected lamina areas are recorded as a region pair, the product of the extension trends of the region pair is calculated, and the absolute value of the difference in the extension trends of the region pair is calculated, and the product of the normalized result of the ratio of the product to the absolute value and the relative position angle of the region pair is recorded as the relevant contribution angle of the region pair; the sum of the relevant contribution angles of all region pairs is recorded as the lamina relevant angle.

[0026] Furthermore, the grayscale enhancement coefficient of the pixel point is obtained according to the similarity between the relative position angle of the suspected lamina area and the lamina-related angle, combined with the extension trend of the suspected lamina area, and the specific steps include the following:

[0027] Any suspected lamina area is recorded as the target suspected lamina area, the minimum absolute value of the difference between the relative position angles of the target suspected lamina area and all other suspected lamina areas is calculated, and the product of the cosine value of the angle corresponding to the minimum value and the extension trend of the target suspected lamina area is recorded as the regional enhancement coefficient of the target suspected lamina area;

[0028] For any pixel point in the target suspected lamina area, the regional enhancement coefficient of the target suspected lamina area is recorded as the grayscale enhancement coefficient of the pixel point; the grayscale enhancement coefficient of the pixel point in the non-suspected lamina area is set to a preset coefficient.

[0029] Furthermore, the ultrasound enhanced image is obtained according to the grayscale enhancement coefficient of the pixel point, the needle center line and the enhancement boundary distance, and the specific steps include the following:

[0030] Lumbar spine ultrasound images The enhanced gray value of each pixel:

[0031]

[0032] In the formula, Lumbar spine ultrasound image The enhanced gray value of each pixel; Lumbar spine ultrasound image The shortest distance from a pixel to the needle centerline; To enhance the demarcation distance; Lumbar spine ultrasound image Gray enhancement coefficient of each pixel; Lumbar spine ultrasound image The gray value of each pixel;

[0033] The gray value of each pixel in the lumbar ultrasound image is replaced with the corresponding enhanced gray value to obtain an ultrasound enhanced image.

[0034] Furthermore, the step of obtaining the needle centerline and enhancement demarcation distance in the lumbar ultrasound image includes the following specific steps:

[0035] Perform edge detection on the lumbar ultrasound image to obtain several edge lines, use a straight line detection algorithm to detect straight lines in the edge lines, and obtain surface lines on both sides of the needle body; record any endpoint of the surface line on either side as a target endpoint, record the distance from the target endpoint to the nearest endpoint of the surface line on the other side as a reference width of the needle body, and record the midpoint from the target endpoint to the nearest endpoint of the surface line on the other side as a midline endpoint of the needle body; according to the method of obtaining the reference width and midline endpoint of the needle body based on the target endpoint, obtain another reference width and another midline endpoint of the needle body from the other endpoint of the surface line where the target endpoint is located, record the average of the two reference widths as the needle width, and record the line connecting the two midline endpoints as the needle midline; half of the needle width is used as the enhanced demarcation distance.

[0036] The present invention also proposes a clinical anesthesia ultrasound image-assisted positioning and guidance system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program stored in the memory to implement the steps of the aforementioned clinical anesthesia ultrasound image-assisted positioning and guidance method.

[0037] The beneficial effects of the technical solution of the present invention are as follows: when the contrast between the needle area and the needle adjacent area is improved, the grayscale value of the pixel point in the needle adjacent area will be reduced, affecting the imaging of the lamina area. According to the characteristics of high grayscale aggregation in the lamina area, the lamina edge probability of the main edge pixel point is obtained, and the lamina area is preliminarily located; according to the characteristics of the lamina area with a high correlation value for the main edge pixel point with a high probability of the lamina edge, the suspected lamina area is obtained, and the regional position where the lamina area may exist is divided; according to the characteristics of the morphological similarity of the lamina area and the similarity of the relative position of the lamina distribution, the grayscale enhancement coefficient of the pixel point is judged, and the possibility that the area where each pixel point is located belongs to the lamina area is accurately judged, thereby improving the enhancement effect of the lumbar ultrasound image. So far, the present invention enhances the lumbar ultrasound image by the grayscale enhancement coefficient of the pixel point, reduces the influence on the imaging of the lamina area in the process of enhancing the lumbar ultrasound image, and improves the auxiliary positioning guidance effect of ultrasound-guided spinal anesthesia. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0039] Figure 1A flowchart of the steps of a clinical anesthesia ultrasound image-assisted positioning and guidance method according to the present invention;

[0040] Figure 2 An ultrasound image of a needle provided in this embodiment;

[0041] Figure 3 This is an ultrasound image of a vertebral plate provided in this embodiment. DETAILED DESCRIPTION

[0042] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of a clinical anesthesia ultrasound image-assisted positioning guidance method and system proposed by the present invention, its specific implementation method, structure, features and effects, in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0043] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0044] The specific scheme of a clinical anesthesia ultrasound image-assisted positioning and guiding method and system provided by the present invention is described in detail below with reference to the accompanying drawings.

[0045] See also Figure 1 , which shows a flowchart of a method for clinical anesthesia ultrasound image-assisted positioning guidance provided by an embodiment of the present invention, the method comprising the following steps:

[0046] Step S001: Acquire the needle centerline and enhancement boundary distance in the lumbar ultrasound image.

[0047] It should be noted that the purpose of this embodiment is to determine the enhancement coefficient of the pixel in the lumbar ultrasound image during ultrasound-guided spinal anesthesia by judging the possibility that the pixel is an anatomical landmark, thereby improving the contrast between the needle area and the needle adjacent area, while avoiding the influence of the enhancement process on the imaging of the anatomical landmark, and achieving auxiliary positioning and guiding puncture. Therefore, the lumbar ultrasound image is first acquired, and in order to divide the needle area and the needle adjacent area, the needle centerline and the enhancement boundary distance are acquired.

[0048] Specifically, the patient is placed in an appropriate position and kept in a lateral or sitting position. The operation area is sterile disinfected and a sterile surgical towel is laid. The probe of the ultrasound device is adjusted to an appropriate frequency, the probe is disinfected, and a sterile ultrasonic coupling agent is applied to the probe. Initial positioning is performed in the patient's lumbar spine area. A paramedian scan is performed at the positioned position, and the angle and pressure of the probe are appropriately adjusted to obtain a lumbar spine ultrasound image.

[0049] It should be further explained that during ultrasound-guided spinal anesthesia, it is mainly necessary to determine the needle position in the lumbar ultrasound image. Since the needle is a slender tubular structure and the material of the needle is relatively hard, high echoes will be generated on both sides of the needle during ultrasound detection. In the lumbar ultrasound image, two brighter lines will be generated on the surface of the needle, while the inside of the needle will not generate echoes, so the inside of the needle appears as low brightness in the image, such as Figure 2 As shown, the needle centerline and enhanced demarcation distance are obtained in the lumbar ultrasound image based on this.

[0050] Specifically, Canny edge detection is performed on the lumbar ultrasound image to obtain several edge lines, and Hough transform is used to detect straight lines in the edge lines to obtain surface lines on both sides of the needle body; any endpoint of the surface line on any side is recorded as a target endpoint, and the distance from the target endpoint to the nearest endpoint of the surface line on the other side is recorded as a reference width of the needle body, and the midpoint from the target endpoint to the nearest endpoint of the surface line on the other side is recorded as a midline endpoint of the needle body; according to the method of obtaining the reference width and midline endpoint of the needle body based on the target endpoint, another reference width and another midline endpoint of the needle body are obtained from the other endpoint of the surface line where the target endpoint is located, the average of the two reference widths is recorded as the needle width, and the line connecting the two midline endpoints is recorded as the needle midline; wherein, Canny edge detection and Hough transform are well-known technologies, and the specific methods are not introduced here;

[0051] Since the needle centerline is located in the middle of the surface lines on both sides of the needle body, half of the needle width is used as the reinforcement demarcation distance.

[0052] It should be noted that when enhancing the lumbar ultrasound image, the grayscale value of the needle area is mainly increased, and the grayscale value of the area adjacent to the needle is reduced, thereby improving the contrast between the needle area and the area adjacent to the needle. Therefore, the needle centerline is used as a reference to determine the enhancement boundary distance and distinguish the needle area from the area adjacent to the needle.

[0053] Step S002: Obtain the main edge pixel points according to the grayscale distribution of the pixel points in the lumbar ultrasound image; obtain the lamina edge probability of the main edge pixel points according to the grayscale values ​​of the main edge pixel points and the pixel points in the lumbar ultrasound image.

[0054] It should be noted that in the process of ultrasound-guided spinal anesthesia, the needle puncture position is adjusted mainly based on the relative position relationship between the needle and the anatomical landmark in the lumbar ultrasound image, where the anatomical landmark is mainly the lamina area. When the needle is enhanced, if the needle is close to the lamina area, when the lamina area is in the vicinity of the needle, if the grayscale value of the lamina area is still reduced, the imaging of the anatomical landmark will be unclear. Therefore, in order to make the relative position relationship between the needle and the anatomical landmark more obvious, the grayscale value of the lamina area should be increased. Therefore, it is necessary to judge the probability that the pixel point in the lumbar ultrasound image is an anatomical landmark. Since the edge between the lamina area and the soft tissue is a strong edge, and the lamina area shows the characteristics of high grayscale aggregation, the probability that the pixel point is the edge of the lamina is first judged based on this.

[0055] It should be further explained that the distinction between regions in the lumbar ultrasound image is mainly based on the change in echo intensity, that is, the change in the grayscale value of the pixel points in the lumbar ultrasound image. Due to the difference in density between the vertebral plate and the soft tissue, there is a strong edge between the vertebral plate area and the soft tissue in the lumbar ultrasound image, so the main edge pixel points in the lumbar ultrasound image are determined.

[0056] Preferably, in one embodiment of the present invention, main edge pixel points are obtained according to the grayscale distribution of pixel points in the lumbar ultrasound image, specifically including: performing Otsu threshold segmentation on the lumbar ultrasound image, and performing binary image hole filling to obtain a lumbar binary image; in the two grayscale areas of the lumbar binary image, the area with the largest average grayscale value is recorded as a high echo area; the edge pixel points of the high echo area are recorded as main edge pixel points; wherein, Otsu threshold segmentation and hole filling are well-known technologies, and the specific methods are not introduced here.

[0057] It should be noted that there is a difference in grayscale value between the lamina area and other adjacent areas, and the edge pixels of the lamina exist in the main edge pixels. The lamina is part of the bone tissue, and the density of the bone is much higher than that of the surrounding soft tissue. Therefore, in the lumbar ultrasound image, the grayscale value of the pixels in the lumbar edge area shows the characteristics of local high grayscale aggregation. The more similar the grayscale values ​​of the pixels on the path from the pixel to the main edge pixel in the lumbar ultrasound image, the smaller the grayscale fluctuation on the path, that is, the greater the correlation value of the pixel in the lumbar ultrasound image to the main edge pixel, the stronger the grayscale aggregation characteristics of the pixel in the lumbar ultrasound image to the main edge pixel. If the grayscale value of the pixel with a large correlation value to the main edge pixel in the lumbar ultrasound image is higher, the local high grayscale aggregation characteristics of the main edge pixel are stronger, and the probability that the main edge pixel belongs to the lamina edge is greater, so the lamina edge probability of the main edge pixel is judged accordingly.

[0058] Preferably, in one embodiment of the present invention, the lamina edge probability of the main edge pixel point is obtained according to the gray value of the main edge pixel point and the pixel point in the lumbar ultrasound image, specifically including:

[0059] Any pixel in the lumbar ultrasound image is recorded as a target pixel. For any main edge pixel, the inverse normalized result of the grayscale variance of all pixels on the straight line from the target pixel to the main edge pixel is recorded as the correlation value of the target pixel to the main edge pixel.

[0060] The product of the normalized result of the correlation value of the target pixel point with respect to the main edge pixel point and the gray value of the target pixel point is recorded as the probability factor of the target pixel point with respect to the main edge pixel point; the sum of the probability factors of all pixels in the lumbar ultrasound image with respect to the main edge pixel point is linearly normalized and recorded as the lamina edge probability of the main edge pixel point.

[0061] It should be noted that since the pixels in the lamina area show the characteristics of local high grayscale aggregation, and the lamina area is adjacent to the main edge pixels, the smaller the grayscale fluctuation on the path from the main edge pixels to the target pixels, that is, the larger the correlation value, the stronger the grayscale aggregation of the pixels on the path from the main edge pixels to the target pixels.

[0062] As an example, the lamina edge probability is formulated as:

[0063]

[0064] In the formula, For the The lamina edge probability of the main edge pixels; is the number of pixels in the lumbar ultrasound image; Lumbar spine ultrasound image The gray value of each pixel; Lumbar spine ultrasound image Pixels for the The correlation value of the main edge pixels; is the normalized exponential function, and the normalized object is all the pixels in the lumbar ultrasound image for the The correlation value of the main edge pixels; is a linear normalization function, and the normalized object is all the main edge pixels. .

[0065] It should be noted that The larger the value is, the more Pixels to The smaller the overall grayscale difference between the pixels of the first main edge pixels, the Pixels to The higher the grayscale concentration of pixels between the first and second main edge pixels, the The gray value of the pixel is used to determine the The greater the weight of the lamina edge probability when there are more main edge pixels.

[0066] Step S003: Classify the pixels in the lumbar ultrasound image according to the correlation values ​​of the pixels to all the main edge pixels and the lamina edge probabilities of all the main edge pixels to obtain the suspected lamina area.

[0067] It should be noted that in the process of grayscale suppression of the pixel points in the vicinity of the needle tip, in order to reduce the impact on the lamina area, it is necessary to determine the position of the lamina area in the lumbar ultrasound image, so the suspected lamina area is acquired. In the lumbar ultrasound image, since the lamina area is distributed near the main edge pixel points, the lamina area has a higher correlation value for the main edge pixel points with a higher probability of the lamina edge. Therefore, according to the correlation value of the pixel points in the lumbar ultrasound image to all the main edge pixel points and the lamina edge probability of all the main edge pixels, the pixels are classified to obtain the suspected lamina area.

[0068] Preferably, in one embodiment of the present invention, according to the correlation values ​​of the pixels in the lumbar ultrasound image with respect to all the main edge pixels and the lamina edge probabilities of all the main edge pixels, the pixels are classified to obtain the suspected lamina area, which specifically includes:

[0069] For any main edge pixel, the product of the lamina edge probability of the main edge pixel and the correlation value of the target pixel to the main edge pixel is recorded as the possible factor of the target pixel to the main edge pixel; the sum of the possible factors of the target pixel to all main edge pixels is recorded as the lamina highlight possibility of the target pixel;

[0070] The pixels in the lumbar ultrasound image are divided into two categories according to the possibility of lamina highlighting using the Otsu threshold method. The pixel points in the category with the greatest possibility of lamina highlighting are recorded as suspected lamina pixels. The area composed of suspected lamina pixels in the lumbar ultrasound image is recorded as suspected lamina area.

[0071] As an example, the lamina highlighting probability is formulated as:

[0072]

[0073] In the formula, Lumbar spine ultrasound image The probability of highlighting the lamina of each pixel is is the number of main edge pixels in the lumbar ultrasound image; Lumbar spine ultrasound image The lamina edge probability of the main edge pixels; Lumbar spine ultrasound image Pixels for the The correlation value of the main edge pixels.

[0074] It should be noted that the first When the correlation value of the pixel point is high for the main edge pixel point with a higher probability of the lamina edge, the The more pixels there are, the greater the possibility that the lamina will be highlighted.

[0075] Step S004: Based on the morphological similarity of the suspected lamina area, the extension trend of the suspected lamina area is obtained; based on the relative position distribution of the suspected lamina area, the relative position angle of the suspected lamina area is obtained, and based on the relative position angle and extension trend of the suspected lamina area, the lamina-related angle is obtained; based on the similarity between the relative position angle of the suspected lamina area and the lamina-related angle, combined with the extension trend of the suspected lamina area, the grayscale enhancement coefficient of the pixel point is obtained.

[0076] It should be noted that in the lumbar ultrasound image, since the main edge pixels correspond to the location with large echo changes, there will be more highlight areas near the main edge pixels, so the analysis based on the characteristics of local high grayscale aggregation can only preliminarily obtain the suspected lamina area. Therefore, further analysis is performed based on the similarity of the morphology of the lamina area and the similarity of the relative distribution position of the lamina. First, the morphological similarity of the suspected lamina area is used to obtain the extension trend of the suspected lamina area, and then the grayscale enhancement coefficient of the pixel point is obtained based on the similarity of the relative distribution position of the lamina combined with the extension trend.

[0077] It should be further explained that in the lumbar ultrasound images, the lamina regions have morphological similarities, the lamina regions appear flat, and the extension directions of the flatness are similar, e.g. Figure 3 As shown, Figure 3 The highlighted areas at L4, L5 and S1 in the figure are lamina areas, so when the extension direction of the suspected lamina area is similar to that of other suspected lamina areas, the suspected lamina area is more likely to be the lamina area.

[0078] Preferably, in one embodiment of the present invention, the extension trend of the suspected lamina region is obtained according to the morphological similarity of the suspected lamina region, which specifically includes:

[0079] The skeleton lines of each suspected lamina area were extracted using a skeleton extraction algorithm; a rectangular coordinate system was established with the lower left corner of the lumbar ultrasound image as the origin, and a straight line was fitted to the skeleton lines of each suspected lamina area to obtain a fitting line. The angle between the fitting line and the y-axis in the clockwise direction was recorded as the main extension angle of the suspected lamina area;

[0080] Any suspected lamina area is recorded as the target suspected lamina area, and the cosine value of the angle corresponding to the absolute value of the difference between the main extension angles of the target suspected lamina area and any other suspected lamina area is recorded as the trend factor between the target suspected lamina area and the suspected lamina area; the average of the trend factors of the target suspected lamina area and all suspected lamina areas is recorded as the extension trend of the target suspected lamina area.

[0081] As an example, extended trending is expressed as:

[0082]

[0083] In the formula, For the The extension trend of the suspected lamina area; is the number of suspected lamina areas; For the The main extension angle of the suspected lamina area; For the The main extension angle of the suspected lamina area; is the absolute value function.

[0084] It should be noted that The greater the extension trend of the suspected lamina area, the The more suspected lamina areas are, the more likely they are to be lamina areas.

[0085] It should be noted that since the lamina is distributed in the posterior part of the vertebrae, the distribution of the lamina area in the lumbar ultrasound image is approximately a straight line. Therefore, the direction of the lamina distribution, that is, the lamina-related angle, is judged in combination with the extension trend of the suspected lamina area to obtain the relative position angle between the suspected lamina areas. If the extension trends of two suspected lamina areas are relatively large and similar, the relative position angle of the two suspected lamina areas is more likely to be the direction of the lamina distribution, thereby improving the accuracy of lamina area acquisition.

[0086] Preferably, in one embodiment of the present invention, the relative position angle of the suspected lamina area is obtained according to the relative position distribution of the suspected lamina area, and the lamina-related angle is obtained according to the relative position angle and extension trend of the suspected lamina area, which specifically includes:

[0087] Obtain the midpoint of the skeleton line of each suspected lamina region; record the angle between the midpoint of the skeleton line of any suspected lamina region and the y-axis in the clockwise direction of the line connecting the midpoint of the skeleton line of any other suspected lamina region as the relative position angle of the two suspected lamina regions;

[0088] Any two suspected lamina areas are recorded as a region pair, the product of the extension trends of the region pair is calculated, and the absolute value of the difference in the extension trends of the region pair is calculated, and the product of the normalized result of the ratio of the product to the absolute value and the relative position angle of the region pair is recorded as the relevant contribution angle of the region pair; the sum of the relevant contribution angles of all region pairs is recorded as the lamina relevant angle.

[0089] As an example, the lamina-related angles are formulated as:

[0090]

[0091] In the formula, is the lamina-related angle; is the number of suspected lamina areas; For the The extension trend of the suspected lamina area; For the The extension trend of the suspected lamina area; For the The suspected lamina area The relative position angle of the suspected lamina area; is the absolute value function; is the normalized exponential function, and the normalized object is all region pairs. .

[0092] It should be noted that The larger the The suspected lamina area The extension trend of the suspected lamina area is larger and similar. The suspected lamina area The larger the relative position angle of the suspected lamina area, the more likely it is the direction of lamina distribution.

[0093] It should be noted that the greater the extension trend of the suspected lamina area and the closer the relative position angle of the suspected lamina area is to the lamina-related angle, the greater the possibility that the suspected lamina area belongs to the lamina area, that is, the larger the regional enhancement coefficient of the suspected lamina area, the larger the grayscale enhancement coefficient of the pixel points in the suspected lamina area.

[0094] Preferably, in one embodiment of the present invention, according to the similarity between the relative position angle of the suspected lamina region and the lamina-related angle, combined with the extension trend of the suspected lamina region, the grayscale enhancement coefficient of the pixel point is obtained, specifically including:

[0095] Any suspected lamina area is recorded as the target suspected lamina area, the minimum absolute value of the difference between the relative position angles of the target suspected lamina area and all other suspected lamina areas is calculated, and the product of the cosine value of the angle corresponding to the minimum value and the extension trend of the target suspected lamina area is recorded as the regional enhancement coefficient of the target suspected lamina area;

[0096] For any pixel point in the target suspected lamina area, the regional enhancement coefficient of the target suspected lamina area is recorded as the grayscale enhancement coefficient of the pixel point; the grayscale enhancement coefficient of the pixel point in the non-suspected lamina area is set to the preset coefficient , This embodiment is described using this as an example.

[0097] As an example, the regional enhancement factor is formulated as:

[0098]

[0099] In the formula, For the The regional enhancement coefficient of the suspected lamina area; For the The extension trend of the suspected lamina area; For the The relative position angle of the suspected lamina area to all other suspected lamina areas; is the lamina-related angle.

[0100] Step S005: Obtain an ultrasound enhanced image and perform auxiliary positioning guidance based on the grayscale enhancement coefficient of the pixel point, the needle center line and the enhancement boundary distance.

[0101] It should be noted that after obtaining the grayscale enhancement coefficient of the pixel point, the needle area and the needle adjacent area need to be enhanced to different degrees. In order to make the needle area more obvious in the image, the grayscale value of the needle area is set to 255. In order to prevent the soft tissue near the needle area from interfering with the imaging of the needle, the grayscale value of the pixel point in the needle adjacent area is reduced. However, when there are lamina area pixels in the needle adjacent area, the lamina area is an anatomical landmark and its grayscale value cannot be reduced, so its grayscale value is increased. For other areas outside the needle area and the needle adjacent area, in order to ensure the original detail information in the lumbar ultrasound image, the original grayscale value is retained unchanged.

[0102] Preferably, in one embodiment of the present invention, an ultrasound enhanced image is obtained according to the grayscale enhancement coefficient of the pixel point, the needle center line and the enhancement boundary distance, specifically including:

[0103] The grayscale enhancement function is constructed based on the grayscale enhancement coefficient of the pixel point, the needle center line and the enhancement boundary distance. The specific construction method is:

[0104]

[0105] In the formula, Lumbar spine ultrasound image The enhanced gray value of each pixel; Lumbar spine ultrasound image The shortest distance from a pixel to the needle centerline; To enhance the demarcation distance; Lumbar spine ultrasound image Gray enhancement coefficient of each pixel; Lumbar spine ultrasound image The gray value of a pixel.

[0106] Furthermore, the grayscale value of each pixel in the lumbar ultrasound image is replaced with the corresponding enhanced grayscale value to obtain an ultrasound enhanced image; and the process of spinal anesthesia is assisted in positioning based on the ultrasound enhanced image.

[0107] At this point, this embodiment is completed.

[0108] The present invention also provides a clinical anesthesia ultrasound image-assisted positioning and guidance system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program stored in the memory to implement the steps of the aforementioned clinical anesthesia ultrasound image-assisted positioning and guidance method.

[0109] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A clinical anesthesia ultrasound image-assisted positioning guidance method, characterized in that: The method comprises the following steps: The needle centerline and enhancement demarcation distance in lumbar ultrasound images were obtained; According to the grayscale distribution of the pixels in the lumbar ultrasound image, the main edge pixels are obtained; according to the grayscale values ​​of the main edge pixels and the pixels in the lumbar ultrasound image, the lamina edge probability of the main edge pixels is obtained; According to the correlation values ​​of the pixels in the lumbar ultrasound image with all the main edge pixels and the lamina edge probabilities of all the main edge pixels, the pixels are classified to obtain the suspected lamina area; According to the morphological similarity of the suspected lamina area, the extension trend of the suspected lamina area is obtained; according to the relative position distribution of the suspected lamina area, the relative position angle of the suspected lamina area is obtained, and according to the relative position angle and extension trend of the suspected lamina area, the lamina-related angle is obtained; according to the similarity between the relative position angle of the suspected lamina area and the lamina-related angle, combined with the extension trend of the suspected lamina area, the grayscale enhancement coefficient of the pixel point is obtained; According to the grayscale enhancement coefficient of the pixel point, the needle centerline and the enhancement boundary distance, an ultrasonic enhanced image is obtained and auxiliary positioning guidance is performed.

2. According to claim 1, a clinical anesthesia ultrasound image-assisted positioning guidance method is characterized in that: The method of obtaining the main edge pixel points according to the grayscale distribution of the pixel points in the lumbar ultrasound image includes the following specific steps: The lumbar ultrasound image is threshold segmented and the binary image hole filling is performed to obtain the lumbar binary image; in the two grayscale areas of the lumbar binary image, the area with the largest average grayscale value is recorded as the high echo area; the edge pixel points of the high echo area are recorded as the main edge pixel points.

3. According to claim 1, a clinical anesthesia ultrasound image-assisted positioning guidance method is characterized in that: The step of obtaining the lamina edge probability of the main edge pixel points according to the grayscale values ​​of the main edge pixel points and the pixel points in the lumbar ultrasound image includes the following specific steps: Any pixel point in the lumbar ultrasound image is recorded as a target pixel point; for any main edge pixel point, the inverse normalization result of the grayscale variance of all pixels on the straight line path from the target pixel point to the main edge pixel point is recorded as the correlation value of the target pixel point to the main edge pixel point; The product of the normalized result of the correlation value of the target pixel point with respect to the main edge pixel point and the gray value of the target pixel point is recorded as the probability factor of the target pixel point with respect to the main edge pixel point; the normalized result of the sum of the probability factors of all pixels in the lumbar ultrasound image with respect to the main edge pixel point is recorded as the lamina edge probability of the main edge pixel point.

4. The clinical anesthesia ultrasound image-assisted positioning guidance method according to claim 1, characterized in that: The specific steps of classifying the pixels according to the correlation values ​​of the pixels in the lumbar ultrasound image to all the main edge pixels and the lamina edge probabilities of all the main edge pixels to obtain the suspected lamina area are as follows: Any pixel in the lumbar ultrasound image is recorded as a target pixel; for any main edge pixel, the product of the lamina edge probability of the main edge pixel and the correlation value of the target pixel to the main edge pixel is recorded as the possible factor of the target pixel to the main edge pixel; the sum of the possible factors of the target pixel to all the main edge pixels is recorded as the lamina highlight possibility of the target pixel; The pixel points in the lumbar ultrasound image are divided into two categories according to the possibility of lamina highlighting using the threshold segmentation method. The pixel points in the category with the greatest possibility of lamina highlighting are recorded as suspected lamina pixels; the area composed of suspected lamina pixels in the lumbar ultrasound image is recorded as suspected lamina area.

5. The clinical anesthesia ultrasound image-assisted positioning guidance method according to claim 1, characterized in that: The step of obtaining the extension trend of the suspected lamina region according to the morphological similarity of the suspected lamina region includes the following specific steps: Extracting skeleton lines of each suspected lamina region using a skeleton extraction algorithm; obtaining a main extension angle of the suspected lamina region according to the direction of the skeleton lines; Any suspected lamina region is recorded as a target suspected lamina region, and the cosine value of the angle corresponding to the absolute value of the difference between the main extension angle of the target suspected lamina region and any other suspected lamina region is recorded as a trend factor between the target suspected lamina region and the suspected lamina region; The average of the trend factors of the target suspected lamina area and all suspected lamina areas is recorded as the extension trend of the target suspected lamina area.

6. The clinical anesthesia ultrasound image-assisted positioning guidance method according to claim 1, characterized in that: The method of obtaining the relative position angle of the suspected lamina region according to the relative position distribution of the suspected lamina region, and obtaining the lamina-related angle according to the relative position angle and extension trend of the suspected lamina region, includes the following specific steps: Obtain the skeleton line and the midpoint of each suspected lamina region; obtain the relative position angle of any two suspected lamina regions according to the midpoints of the skeleton lines of the two suspected lamina regions; Any two suspected lamina areas are recorded as a region pair, the product of the extension trends of the region pair is calculated, and the absolute value of the difference in the extension trends of the region pair is calculated, and the product of the normalized result of the ratio of the product to the absolute value and the relative position angle of the region pair is recorded as the relevant contribution angle of the region pair; the sum of the relevant contribution angles of all region pairs is recorded as the lamina relevant angle.

7. The clinical anesthesia ultrasound image-assisted positioning guidance method according to claim 1, characterized in that: The grayscale enhancement coefficient of the pixel point is obtained according to the similarity between the relative position angle of the suspected lamina area and the lamina related angle, combined with the extension trend of the suspected lamina area, and the specific steps include the following: Any suspected lamina area is recorded as the target suspected lamina area, the minimum absolute value of the difference between the relative position angles of the target suspected lamina area and all other suspected lamina areas is calculated, and the product of the cosine value of the angle corresponding to the minimum value and the extension trend of the target suspected lamina area is recorded as the regional enhancement coefficient of the target suspected lamina area; For any pixel point in the target suspected lamina area, the regional enhancement coefficient of the target suspected lamina area is recorded as the grayscale enhancement coefficient of the pixel point; the grayscale enhancement coefficient of the pixel point in the non-suspected lamina area is set to a preset coefficient.

8. The clinical anesthesia ultrasound image-assisted positioning guidance method according to claim 1, characterized in that: The specific steps of obtaining the ultrasound enhanced image according to the grayscale enhancement coefficient of the pixel point, the needle center line and the enhancement boundary distance are as follows: Lumbar spine ultrasound images The enhanced gray value of each pixel: In the formula, Lumbar spine ultrasound image The enhanced gray value of each pixel; Lumbar spine ultrasound image The shortest distance from a pixel to the needle centerline; To enhance the demarcation distance; Lumbar spine ultrasound image Gray enhancement coefficient of each pixel; Lumbar spine ultrasound image The gray value of each pixel; The gray value of each pixel in the lumbar ultrasound image is replaced with the corresponding enhanced gray value to obtain an ultrasound enhanced image.

9. The clinical anesthesia ultrasound image-assisted positioning guidance method according to claim 1, characterized in that: The specific steps of obtaining the needle centerline and enhancement demarcation distance in the lumbar ultrasound image are as follows: Perform edge detection on the lumbar ultrasound image to obtain several edge lines, use a straight line detection algorithm to detect straight lines in the edge lines, and obtain surface lines on both sides of the needle body; record any endpoint of the surface line on either side as a target endpoint, record the distance from the target endpoint to the nearest endpoint of the surface line on the other side as a reference width of the needle body, and record the midpoint from the target endpoint to the nearest endpoint of the surface line on the other side as a midline endpoint of the needle body; according to the method of obtaining the reference width and midline endpoint of the needle body based on the target endpoint, obtain another reference width and another midline endpoint of the needle body from the other endpoint of the surface line where the target endpoint is located, record the average of the two reference widths as the needle width, and record the line connecting the two midline endpoints as the needle midline; half of the needle width is used as the enhanced demarcation distance.

10. A clinical anesthesia ultrasound image-assisted positioning guidance system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the computer program is executed by the processor, the steps of a clinical anesthesia ultrasound image-assisted positioning and guidance method as described in any one of claims 1-9 are implemented.

Images