A WEB-based ultrasonic image intelligent acquisition system and method

Abstract

The present invention discloses a WEB-based ultrasonic image intelligent acquisition system and method, which relates to the field of computer technology. The technical solution mainly includes an image acquisition module, an analysis module, a first processing module, a second processing module, and an output module, which are used to perform color re-adjustment processing on lesion data under different classifications to obtain multi-level atlas sharing data. Through feature extraction and multi-level lesion analysis, accurate positioning and refined analysis of lesion areas are achieved. Feature extraction is performed on qualified image data and unqualified image data respectively to generate feature data A and feature data B. By calculating the density and distribution of dark and light pixel points, the system can accurately locate the lesion area, and generate one to four levels of lesion data based on these data, thereby achieving multi-level analysis of the lesion area. The refined analysis method not only helps medical staff to have a deeper understanding of the characteristics and progress of the lesion.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and more specifically, to a WEB-based ultrasonic image intelligent acquisition system and method. Background Art

[0002] WEB-based intelligent ultrasound image acquisition technology can automatically analyze ultrasound images, and further combine with the WEB's integrated platform for remote acquisition, transmission and analysis, allowing different medical institutions to share patient data, promote multidisciplinary collaborative diagnosis and treatment, realize remote office and team collaboration, and make geographical location no longer an obstacle to cooperation. In the existing technology, during the remote acquisition stage of ultrasound images, the collected ultrasound images will be preliminarily screened, and unqualified images will be optimized to ensure the quality of the input data. Then, on the WEB integrated collaborative diagnosis and treatment platform, combined with the WEB platform's analysis system, collaborative diagnosis and treatment will be carried out.

[0003] Doctors or technicians in different geographical locations can remotely control ultrasound equipment for image acquisition through a WEB platform. Generally, the direction and pressure of the ultrasound probe are controlled through network operations, and features are extracted from the image to identify key image features, such as brightness, darkness, texture, edges, etc. However, in actual operations, the causes of the same lesion are different. For example, in the patient's abdominal digestive system, gallstone bleeding and liver cyst bleeding are both bleeding, which appear as dark colors with weak reflections under ultrasound image development. The acquisition modules of the existing technology can often only highlight image features of the same color system, and cannot deeply distinguish the causes of image features of the same color system, and cannot accurately capture the data features between the lesion area and normal tissue. Summary of the invention

[0004] In view of the deficiencies in the prior art, the purpose of the present invention is to provide a WEB-based ultrasonic image intelligent acquisition system and method, aiming to solve the above technical problems.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A WEB-based ultrasonic image intelligent acquisition system, comprising:

[0007] An image acquisition module is used to obtain multiple sets of image data and screen the multiple sets of image data to obtain qualified image data and unqualified image data;

[0008] Analysis module, used to analyze qualified image data to obtain feature data , analyze the unqualified image data to obtain feature data ;

[0009] The first processing module is used to process the feature data With feature data Perform recombinant contrast processing to obtain dark data of the lesion area and bright color data of the lesion area ;

[0010] The second processing module is used to process the dark data of the lesion area The size of is assigned to obtain the primary lesion data; according to the bright color data of the lesion area The size of is assigned to obtain the secondary lesion data; according to the dark data of the lesion area and bright color data of the lesion area The distance between the two edges is assigned to obtain the third-level lesion data; according to the dark data of the lesion area , bright color data of lesion area The area with overlapping edges was assigned values ​​to obtain the fourth-level lesion data;

[0011] The output module is used to perform color re-adjustment processing on the lesion data under different grades to obtain multi-level atlas shared data.

[0012] As a further solution of the present invention: analyzing the qualified image data to obtain characteristic data ,include:

[0013] For each set of image data in multiple sets of images,

[0014] Processing to obtain feature data ,in, Represents the contrast of ultrasound images. The calculation formula is:

[0015] ;

[0016] in, Represents an ultrasound image, Represents ultrasound image In Location The pixel value of Represents ultrasound image The average pixel value, Represents ultrasound image The total number of pixels in and Represents ultrasound images In Location The horizontal and vertical gradients at , , Represents the weight coefficient.

[0017] As a further solution of the present invention: With feature data Perform recombinant contrast processing to obtain dark data of the lesion area Bright color data of lesion area ,include:

[0018] Confirming the ultrasound image The collection of dark pixels is set as , preset the threshold of dark pixels , when the gray value of the pixel is less than the threshold of the dark pixel When , the pixel is considered as a dark pixel and added to the set In the ultrasound image, The total number of pixels is set as , according to the set of dark pixels , determine the number of dark pixels and set it as , the number of dark pixels Divide by the total number of pixels , get the dark data of the lesion area ;

[0019] Confirming the ultrasound image The set of medium and bright pixels is set as , preset the threshold of bright pixels , when the gray value of the pixel is greater than the threshold of the bright pixel , the pixel is considered a bright pixel and added to the set In the ultrasound image, The total number of pixels is set as , according to the set of bright pixels , determine the number of bright pixels and set it , the number of bright pixels Divide by the total number of pixels , get the bright color data of the lesion area .

[0020] As a further solution of the present invention: according to the dark data of the lesion area , bright color data of lesion area The distance between the two edges is assigned to obtain the third-level lesion data, including:

[0021] Determine each pixel in the dark pixel set D And calculate the edge pixels , edge pixels The calculation formula is: ;

[0022] Determine the bright pixel set Each pixel in And calculate the edge pixels , edge pixels The calculation formula is: ;

[0023] in, and Represents pixel points The horizontal and vertical gradient values, and Represents pixel The horizontal gradient and vertical gradient values, and Represents pixel , The multi-level gradient processing value of Location The calculation formula is:

[0024] ;

[0025] in, represents the Gaussian kernel, represents the convolution operation, Represents an ultrasound image, Indicates the zoom ratio, Represents the arc direction angle of the pixel under the horizontal gradient and vertical gradient, which is obtained by dividing the horizontal gradient by the vertical gradient. Represents the statistical processing of the histogram function, Indicates contrast enhancement processing, by obtaining pixel points With pixels The pixel values ​​of the two edges are used to calculate the distance between the two edges. , and Divide by the number of pairs of dark pixels and light pixels to get the average distance , the number of pairings is the total number of groups of total distance, and the average distance obtained under different numbers of groups is Each set of ultrasound images is obtained by assigning values ​​according to the numerical values. The data of grade 3 lesions.

[0026] As a further solution of the present invention: according to the ultrasound image Medium dark pixel collection The data is at the preset dark pixel threshold Based on , which is used to further subdivide the dark areas at different levels, where the pixel brightness is The following settings are for the first level dark area 1. Pixel brightness is Above and in The following settings are for the secondary dark area 2. Pixel brightness The above settings are for the third-level dark area 3. For each subdivided dark area, extract the corresponding dark feature vector , The calculation formula is:

[0027] ;

[0028] in, Indicates the number of pixels in dark areas of different levels, For the coordinates The gray value of the enhanced image at Indicates The average gray value of the dark area is calculated based on the dark feature vectors of the pixels in the dark areas at different levels. Arrange them according to the numerical values, and assign values ​​according to the relationship between the numerical values ​​to obtain the primary lesion data.

[0029] As a further solution of the present invention: according to the size of the primary lesion data, the calculation formula is adjusted according to the color system Processing, and combining pseudo color technology to generate different color areas, setting multiple readjustment thresholds , used to subdivide the hierarchical data under different rescaling treatments;

[0030] in The value size is The following settings are for the green edge area. The value size is Above and in The following settings are the yellow edge area. The value size is The above settings are for the red edge area, and the color system is adjusted by the calculation formula for:

[0031] ;

[0032] in, Indicates the number of pixels in dark areas of different levels, , These are color re-adjustment parameters. Represents the average grayscale value of all pixels in set D, which is used to reflect the overall brightness level of the pixel group currently being processed.

[0033] As a further solution of the present invention: according to the size of the primary lesion data, the calculation formula is adjusted according to the color system Processing is performed and different color areas are generated in combination with pseudo color technology. The pseudo color technology includes a data acquisition unit and a processing unit, wherein the data acquisition unit further includes a signal conversion module and a scanning module, the signal conversion module is responsible for converting analog signals into digital signals, and the scanning module is used to scan image data of different dark areas. The processing unit further includes an image reconstruction module and a WEB sharing module, the image reconstruction module is used to perform B-mode imaging processing on digital signals, and the WEB sharing module is used to transmit the acquired digital signals to the WEB platform for remote synchronous processing.

[0034] A WEB-based ultrasonic image intelligent acquisition method comprises the following steps:

[0035] S1: acquiring image data through an image acquisition module, and screening the image data to obtain qualified image data and unqualified image data;

[0036] S2: Analyze qualified image data to obtain feature data ; Used to analyze unqualified image data to obtain feature data ;

[0037] S3: By analyzing the feature data With feature data Perform recombinant contrast processing to obtain dark data of the lesion area , bright color data of lesion area ;

[0038] S4: Multi-level lesion data is obtained by assigning values ​​to the size of the dark data and the light data of the lesion area, as well as the distance between the two edges and the size of the overlapping area, and color re-adjustment is performed to obtain multi-level atlas shared data.

[0039] As a further solution of the present invention: in step S1, it also includes remotely operating an ultrasonic device to collect images, and the image collection module is an ultrasonic probe and an ultrasonic scanner, wherein the ultrasonic probe is used to generate and receive ultrasonic signals, and the ultrasonic scanner is used to control signal amplification, filtering, and digital image reconstruction;

[0040] Step S2 also includes evaluating the quality of the ultrasound image and triggering a re-acquisition instruction when the quality is lower than a predetermined standard. The image acquisition module also includes a remote control unit for remotely operating the ultrasound device to acquire images.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] 1. The WEB-based ultrasound image intelligent acquisition system significantly improves the diversity and reliability of image data through multi-source data collection and quality control, allowing medical staff in different regions to upload multiple sets of ultrasound image data. After structural similarity screening, the quality of images entering the analysis stage is ensured. It not only increases the diversity of image samples, but also effectively excludes unqualified images caused by improper operation or equipment problems, providing a solid foundation for subsequent lesion area analysis. Furthermore, on this basis, the system calculates the uniformity, edge clarity and contrast of the image, and assigns corresponding weight coefficients to form a comprehensive score, thereby achieving a comprehensive evaluation of image quality.

[0043] 2. Through feature extraction and multi-level lesion analysis, accurate positioning and refined analysis of the lesion area are achieved, and feature extraction is performed on qualified image data and unqualified image data to generate feature data and feature data ,The dual-track feature extraction method can provide a more comprehensive understanding of the information in the image, especially the information of the lesion area. By calculating the density and distribution of dark and light pixels, the system can accurately locate the lesion area, and generate one to four level lesion data based on these data, realizing a multi-level analysis of the lesion area.

[0044] 3. Through pseudo-color technology and color re-adjustment methods, different levels of lesion data are mapped to different colors, such as red, yellow, and green, to generate multi-level atlas shared data, thereby improving the readability of the image. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. 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 creative work.

[0046] Figure 1 It is a schematic diagram of a WEB-based ultrasonic image intelligent acquisition system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0048] Reference Figure 1The present invention further describes an embodiment of a WEB-based ultrasonic image intelligent acquisition system and method.

[0049] A WEB-based ultrasonic image intelligent acquisition system, comprising:

[0050] An image acquisition module is used to obtain multiple sets of image data and screen the multiple sets of image data to obtain qualified image data and unqualified image data;

[0051] Analysis module, used to analyze qualified image data to obtain feature data ; Used to analyze unqualified image data to obtain feature data ;

[0052] The first processing module is used to process the feature data With feature data Perform recombinant contrast processing to obtain dark data of the lesion area , bright color data of lesion area ;

[0053] The second processing module uses a preset algorithm to calculate the dark data of the lesion area. The size of the lesion is assigned to obtain the primary lesion data; according to the bright color data of the lesion area The size of the lesion is assigned to obtain the secondary lesion data; according to the dark data of the lesion area , bright color data of lesion area The distance between the two edges is assigned to obtain the third-level lesion data; according to the dark data of the lesion area , bright color data of lesion area The area with overlapping edges was assigned values ​​to obtain the fourth-level lesion data;

[0054] The output module is used to perform color re-adjustment processing on the lesion data under different grades to obtain multi-level atlas shared data.

[0055] In view of the problem in the prior art that in the patient's abdominal digestive system, both gallstone bleeding and liver cyst bleeding are bleeding, and both appear as dark colors with weak reflections under ultrasonic image development, and the acquisition module of the prior art can often only highlight image features of the same color system, and cannot deeply distinguish the triggering factors of image features of the same color system, and cannot accurately capture the data features between the diseased area and normal tissue, the above-mentioned technical solution is now adopted to solve the problem. The above-mentioned technical solution is mainly composed of an image acquisition module, an analysis module, a first processing module, a second processing module, and an output module.

[0056] The configured image acquisition module is an acquisition module connected to a WEB data sharing platform in the prior art, which is used to obtain multiple groups of ultrasonic image data and screen them to obtain qualified image data and unqualified image data. In combination with the WEB sharing platform, medical staff in different areas can take ultrasonic images of the same patient, and adjust the shooting angle according to the processing method of each medical staff, so as to generate multiple groups of ultrasonic image data at the patient's abdominal digestive system. In each group of ultrasonic image data, the medical staff adjusts and shoots to mark the areas that need to be focused on and the areas that do not need to be focused on. The image acquisition module classifies them according to different medical staff, and further classifies them according to the structural similarity index based on the multiple groups of annotated ultrasonic images, and classifies the key detection areas with the most annotations as qualified image data, and conversely, the areas that do not need to be focused on are classified as unqualified image data. The ultrasonic images annotated by multiple groups of medical staff are further classified according to the structural similarity index. To measure the similarity between each image, compare the similarity between the areas that need to be focused on and the areas that do not need to be focused on in the image, classify the areas that need to be focused on as qualified image data, and classify the areas that do not need to be focused on as unqualified image data. and is the average value of image x and y, and Represent the variance of images x and y respectively, represents the covariance of images x and y, , is a small constant to avoid zero denominator.

[0057] The configured analysis module calculates the uniformity, edge clarity and contrast of each image, and calculates the comprehensive score of each image according to the preset weight coefficient to obtain feature data. , characteristic data , and the first processing module and the second processing module are configured to process the feature data , characteristic data Further processing is carried out, and the data are classified into level one, level two, level three, and level four lesion data according to the position of each feature point in the image. The dark area blocks in the lesion area are color-recalibrated in combination with the output module to obtain multi-level atlas sharing data. The multi-level atlas is classified into three colors: red, yellow, and green according to the severity of the lesion and the difference between gallbladder stone bleeding and liver cyst bleeding, so as to facilitate timely detection and distinction by the WEB sharing platform in different areas. Different from traditional shared data, after being processed by the intelligent acquisition system, the WEB sharing platform can be used to enable medical staff in different areas to take ultrasonic images of the same patient. According to the processing method of each medical staff, the shooting angle is adjusted to generate multiple groups of different ultrasonic images of the same patient and the same position taken by different medical staff. Finally, the dark pixels and bright pixels of multiple groups of different ultrasonic images are analyzed, and the positions and coverage areas of the pixels are compared to accurately capture the data characteristics between the lesion area and the normal tissue, and automatically distinguish gallbladder stone bleeding from liver cyst bleeding.

[0058] like Figure 1 In the above, multiple sets of image data are obtained through the image acquisition module, that is, multiple sets of ultrasound images are obtained. After the data is collected, it is necessary to screen multiple sets of image data to obtain qualified image data and unqualified image data, which is specifically as follows:

[0059] Analyze qualified image data to obtain feature data ,include:

[0060] For each set of image data in multiple sets of images,

[0061] Processing to obtain feature data ,in, Represents the contrast of ultrasound images. The calculation formula is:

[0062] ;

[0063] in, Represents an ultrasound image, Represents ultrasound image In Location The pixel value of Represents ultrasound image The average pixel value, Represents ultrasound image The total number of pixels in and Represents ultrasound images In Location The horizontal and vertical gradients at , , Represents the weight coefficient.

[0064] The specific process of processing each set of image data in the multiple sets of images is as follows: first, multiple sets of ultrasound images are acquired by the image acquisition module. That is, through the WEB platform, under the role of remote control, medical staff in different regions can control the ultrasound image capture end to capture the abdomen of the same patient. For medical staff in different fields, the focus areas and angles of their capture must be different, so multiple sets of different ultrasound images will be obtained. The ultrasound image It is displayed in grayscale, that is, the gradient from black to white represents different tissue densities or reflection intensities. According to different tissue densities or reflection intensities, the image is feature extracted to screen out ultrasound images with different features such as brightness, darkness, texture, and edges. , so next, for each set of ultrasound images Analyze to obtain characteristic data . Characteristic data It consists of three parts. The first part is the weight coefficient. Multiply by the sum of the absolute values ​​of the differences between the pixel values ​​of all pixels and their mean, and divide by the total number of pixels , the second part is the weight coefficient Multiply by the sum of the absolute values ​​of the horizontal and vertical gradients of all pixels. The third part is the weight coefficient Multiply the contrast of the ultrasound image , the feature data A obtained by the above calculation can be used to determine which images are qualified images and which are unqualified images according to certain thresholds or rules. The weight coefficient set Indicates the importance of image brightness information. The larger its value, the more important the overall brightness of the image is, and vice versa. The weight coefficient set Indicates the importance of image texture information. The larger its value, the more important the local details of the image are, and vice versa. The weight coefficient set Indicates the importance of image contrast information. The larger its value is, the greater the difference between different areas of the image, and vice versa.

[0065] Similarly, qualified image data can also be analyzed to obtain feature data ,include:

[0066] For each set of image data in multiple sets of images,

[0067] Processing to obtain feature data ,in, represents the contrast of ultrasound image, where The calculation formula is:

[0068] ;

[0069] in, Represents an ultrasound image, Represents ultrasound image In Location The pixel value of Represents ultrasound image The average pixel value, Represents ultrasound image The total number of pixels in and Represents ultrasound images In Location The horizontal and vertical gradients at , , Represents the weight coefficient.

[0070] like Figure 1 As mentioned above, the feature data With feature data Perform recombinant contrast processing to obtain dark data of the lesion area Bright color data of lesion area ,include:

[0071] Confirming the ultrasound image The collection of dark pixels is set as , preset the threshold of dark pixels , when the gray value of the pixel is less than the threshold of the dark pixel When , the pixel is considered as a dark pixel and added to the set In the ultrasound image, The total number of pixels is set as , according to the set of dark pixels , determine the number of dark pixels and set it as , the number of dark pixels Divide by the total number of pixels , get the dark data of the lesion area ;

[0072] Confirming the ultrasound image The set of medium and bright pixels is set as , preset the threshold of bright pixels , when the gray value of the pixel is greater than the threshold of the bright pixel , the pixel is considered a bright pixel and added to the set In the ultrasound image, The total number of pixels is set as , according to the set of bright pixels , determine the number of bright pixels and set it , the number of bright pixels Divide by the total number of pixels , get the bright color data of the lesion area .

[0073] Specifically, the first step is to set the threshold of dark-tone pixels. and the threshold of bright-tone pixels , the second step is based on the threshold and , classify each pixel in the image as dark or light, and the third step is to calculate the number of dark pixels and the number of bright-tone pixels , and finally use the total number of pixels To calculate the density of dark and light pixels and , can quickly identify and quantitatively analyze abnormal areas in images. When diagnosing bleeding gallstones and hepatic cysts, abnormal areas often appear different in color or texture from the surrounding healthy tissue. In this way, quantitative indicators such as the number and density of pixels are provided, making the analysis results more objective and repeatable. It can be applied to various types of ultrasound images by simply adjusting the threshold appropriately, and can also automatically identify and quantify abnormal areas in images, reducing the need for manual intervention.

[0074] The threshold value is set and Adjust according to dark and light tones, specifically, for the dark tone pixel threshold Since gallstone bleeding usually occurs in the gallbladder, its dark area will be limited to the anatomical position of the gallbladder, and because it is caused by stones, it may be accompanied by a stone shadow with strong echoes. There is often an acoustic shadow behind the stone, that is, a dark band appears in the area behind the stone because the sound waves are blocked. The boundary clarity of the bleeding area is lower than that of liver cyst bleeding, and the blood can spread to the surrounding tissues. If real-time observation is performed, the blood can move in the gallbladder as the patient's position changes. When setting the threshold, the dark tone threshold can be ,in It is an empirical constant, usually with a value of 1 to 3. For liver cysts with bleeding, the location is inside the liver. Before bleeding, the cyst usually appears as a clear-bordered, uniformly anechoic area. After bleeding, the echo inside the cyst is enhanced, and there may be dot-shaped or mass-shaped high-echo areas, representing blood clots. Therefore, there will be a change state. In addition, the border of the bleeding cyst is relatively clear, which is manifested as the border clarity of the bleeding area is higher than that of gallstone bleeding. Therefore, the bright tone threshold can be selected as ,in is an empirical constant, usually ranging from 1 to 3. Represents the average gray value, which is the average gray value of all pixels in the image and reflects the overall brightness of the entire image. It represents the standard deviation, which is used to measure the discreteness of the grayscale value of the image. The larger the standard deviation, the more dispersed the grayscale value distribution of the image is, and the smaller the standard deviation, the more concentrated the grayscale value distribution is.

[0075] like Figure 1 As shown, further, the dark data of the lesion area can be , bright color data of lesion area The distance between the two edges is assigned to obtain the third-level lesion data, including:

[0076] Determine each pixel in the dark pixel set D And calculate the edge pixels , edge pixels The calculation formula is: ;

[0077] Determine the bright pixel set Each pixel in And calculate the edge pixels , edge pixels The calculation formula is: ;

[0078] in, and Represents pixel points The horizontal and vertical gradient values, and Represents pixel The horizontal gradient and vertical gradient values, and Represents pixel , The multi-level gradient processing value of Location The calculation formula is:

[0079] ;

[0080] in, represents the Gaussian kernel, represents the convolution operation, Represents an ultrasound image, Indicates the zoom ratio, It represents the arc direction angle of the pixel point under the horizontal gradient and vertical gradient, which is obtained by dividing the horizontal gradient by the vertical gradient. With pixels The pixel values ​​of the two edges are used to calculate the distance between the two edges. , and Divide by the number of pairs of dark pixels and light pixels to get the average distance , the number of pairings is the total number of groups of total distance. The average distance obtained under different numbers of groups Each set of ultrasound images is obtained by assigning values ​​according to the numerical values. The data of grade 3 lesions.

[0081] The specific calculation process is:

[0082] First, the input image Using Gaussian kernel Perform convolution operation, cover each pixel on the ultrasound image by moving the Gaussian kernel, align the center of the Gaussian kernel with the current pixel, and then calculate the weighted sum of the Gaussian kernel and the corresponding area of ​​the image for smoothing to reduce the impact of noise;

[0083] The convolution result is then passed to the histogram function , through the histogram function Statistical processing is performed to depict the distribution of the data in order to redistribute the pixel values ​​of the image to make it more suitable for subsequent processing steps; finally, the result of the histogram operation is multiplied by a scaling factor , scaling It can be adjusted as needed to accommodate deviations at different shooting positions.

[0084] Specifically for ultrasound images, it can be understood as identifying and quantifying the lesion area by analyzing the dark-tone pixels in the ultrasound image. First, a Gaussian filter is applied to the ultrasound image to generate a smoothed image, and then contrast enhancement is performed. Histogram equalization is applied to the smoothed image to generate a contrast-enhanced image, and the contrast-enhanced image is interpolated and amplified to generate a high-resolution image. .

[0085] Describe the distribution of data to redistribute the pixel values ​​of the image. Specifically, through histogram equalization, the contrast of the image is enhanced by adjusting the distribution of different brightness levels in the image, so that the details in the image are clearer, and the histogram of the original image is adjusted to match it.

[0086] When implementing these two processes, the number of pixels at each gray level in the ultrasound image is first calculated to form a histogram. Then, the cumulative distribution function of each gray level is calculated and the distribution function is converted into a uniform distribution. Finally, according to the adjusted cumulative distribution function, the gray value of each pixel is mapped to a new value and the image is reconstructed, thereby optimizing the contrast and improving the visual effect and recognizability of the image.

[0087] The multi-level gradient processing numerical value is specifically to identify and quantify the lesion area by analyzing the dark-tone pixels in the ultrasound image. First, the ultrasound image Applying a Gaussian filter to generate a smoothed image, performing contrast enhancement, applying histogram equalization to the smoothed image to generate a contrast enhanced image, and performing interpolation and amplification on the contrast enhanced image to generate a high-resolution image , , calculate high-resolution images The gradient in the horizontal direction , calculate high-resolution images The vertical gradient , calculate the edge pixels of dark pixels , set It represents the arc direction angle of the pixel under the horizontal gradient and vertical gradient, which is obtained by dividing the horizontal gradient by the vertical gradient, that is, according to its gradient direction Determine adjacent pixels, such as scanning a certain pixel point The gradient value at , , then the gradient direction for , Indicates the scaling ratio. In actual operation, it is used to assign values ​​to ultrasound images at each scale to ensure that ultrasound images at each scaling ratio can be processed through multi-level gradient processing. Each pixel in , its edge pixels The calculation principle is the same as that of each pixel in the dark pixel set D The edge pixels The same is used to obtain the distances between different dark areas in the same ultrasound image.

[0088] By getting the pixel With pixels The pixel values ​​of the two edges are used to calculate the distance between the two edges. , and Divide by the number of pairs of dark pixels and light pixels to get the average distance , the number of pairings is the total number of groups of total distance, and the average distance obtained under different numbers of groups is Each set of ultrasound images is obtained by assigning values ​​according to the numerical values. The third-level lesion data can be obtained after being processed in this step. The distances between different dark areas and the distances between different light areas under each set of ultrasound images can be obtained. As mentioned above, the dark areas can be traced back to the bleeding location, and the calculated distance of the dark areas is the distance for tracing back the bleeding location, which is used to determine whether the bleeding is from the gallbladder or the liver cyst in the abdomen, and the calculated distance of the light areas is the distance for tracing back the location of the stones. The stones in the gallbladder will appear as strong echo points in the ultrasound image, with obvious acoustic shadows behind them, and the vascular wall, fibrous tissue or scar tissue usually appear as medium-intensity echoes in the ultrasound image, but relative to the surrounding soft tissues, they still appear relatively bright, so it can be determined whether the bleeding location of the gallbladder or the liver cyst is a stone or a tumor, that is, each set of ultrasound images The values ​​reflected by the third-level lesion data can determine whether the bleeding location of the gallbladder and liver cyst is a stone or a tumor and the distance to the stone location.

[0089] like Figure 1 As described above, in determining whether the bleeding location of the gallbladder or liver cyst is a stone or a tumor and the distance of the stone location, according to the ultrasound image Medium dark pixel collection After the data is collected, some types of tumors, especially solid tumors, usually have strong echoes due to dense cells and appear bright. In the acute inflammatory area, due to cell infiltration and edema, the echoes may also be enhanced and appear bright. Further, blood clots formed after bleeding may appear as medium-intensity echoes with uneven distribution, and may also appear bright. The difference between them lies in the degree of brightness. For dark pixels, after blood accumulates, especially in the early stages, the blood composition is relatively uniform and the density is low, so the ability to reflect sound waves is weak. Therefore, it appears as a low-echo area in the ultrasound image, that is, it appears dark. Necrotic tissue loses its normal cell structure and function, and has a weak ability to reflect sound waves. Therefore, it appears as a low-echo area in the ultrasound image. In addition, if there is a cyst in the gallbladder, the liquid area in the cyst appears as an echo-free area in the ultrasound image, which also appears dark. The difference between them also lies in the degree of darkness.

[0090] Therefore, for the dark area, the preset dark pixel threshold Based on this, multiple thresholds are set , which is used to further subdivide the dark areas at different levels, where the pixel brightness is The following settings are for the first level dark area 1. Pixel brightness is Above and in The following settings are for the secondary dark area 2. Pixel brightness The above settings are for the third-level dark area 3. For each subdivided dark area, extract the corresponding dark feature vector , to further distinguish the bleeding differences under different levels of dark areas, and their corresponding The calculation formula is:

[0091] ;

[0092] in, Indicates the number of pixels in dark areas of different levels, For the coordinates The gray value of the enhanced image at Indicates The average gray value of the dark area at different levels is calculated based on the dark feature vectors of the pixels in the dark areas at different levels. Arrange them according to the numerical size, and assign values ​​according to the relationship between the numerical sizes to obtain the primary lesion data. Through the quantitative analysis of the dark areas in the ultrasound image, it helps to distinguish different types of lesions. By calculating the average gray value and standard deviation of each dark area, information about the brightness and brightness variability of the area can be obtained. This information can help doctors better understand the nature of the lesion and distinguish between hemorrhage, cysts, necrotic tissue, etc. Medium light pixel collection The data can also be used to set multiple thresholds based on the preset bright pixel threshold to further subdivide bright areas of different levels and extract corresponding bright feature vectors respectively.

[0093] like Figure 1 As described above, further, taking the dark area as an example, after distinguishing hemorrhage, cyst, and necrosis, since the difference in the ultrasound image is not great, only the degree of darkness is different, the calculation formula can be adjusted according to the color system based on the size of the primary lesion data. Processing, and combining pseudo color technology to generate different color areas, setting multiple readjustment thresholds , used to subdivide the hierarchical data under different rescaling treatments.

[0094] in The value size is The following settings are for the green edge area. The value size is Above and in The following settings are the yellow edge area. The value size is The above settings are for the red edge area, and the color system is adjusted by the calculation formula for:

[0095] ;

[0096] in, Indicates the number of pixels in dark areas of different levels, , These are color re-adjustment parameters. Represents the average grayscale value of all pixels in set D, which is used to reflect the overall brightness level of the pixel group currently being processed. The re-adjustment calculation formula Color rescaling parameters in For ultrasound images The magnification factor in is used to change the magnification of the ratio of the original pixel intensity to the standard deviation, thereby affecting the size of the final output Er value. It will increase the contrast between different dark areas and highlight the difference between the lesion area and the surrounding tissue.

[0097] The recalculation formula Color rescaling parameters in For ultrasound images The offset in is used to adjust the brightness level of the entire image. Values ​​brighten the image overall, negative A lower value will darken the image to highlight the difference between the dark area and the surrounding tissue. Certain features or areas are highlighted by assigning specific colors to different grayscale values. Important areas or areas that require special attention are marked in red, while relatively safe or less important areas are represented in green. This helps to quickly identify key information in the image, especially in the process of multi-person and multi-angle processing, and can better reflect the importance of multi-color readjustment.

[0098] like Figure 1 As shown, according to the size of the primary lesion data, the color system is adjusted according to the calculation formula Processing is performed and different color areas are generated in combination with pseudo color technology. The pseudo color technology includes a data acquisition unit and a processing unit, wherein the data acquisition unit further includes a signal conversion module and a scanning module, the signal conversion module is responsible for converting analog signals into digital signals, and the scanning module is used to scan image data of different dark areas. The processing unit further includes an image reconstruction module and a WEB sharing module, the image reconstruction module is used to perform B-mode imaging processing on digital signals, and the WEB sharing module is used to transmit the acquired digital signals to the WEB platform for remote synchronous processing.

[0099] A WEB-based ultrasonic image intelligent acquisition method comprises the following steps:

[0100] S1: acquiring image data through an image acquisition module, and screening the image data to obtain qualified image data and unqualified image data;

[0101] S2: Analyze qualified image data to obtain feature data ; Used to analyze unqualified image data to obtain feature data ;

[0102] S3: By analyzing the feature data With feature data Perform recombinant contrast processing to obtain dark data of the lesion area , bright color data of lesion area ;

[0103] S4: Multi-level lesion data is obtained by assigning values ​​to the size of the dark data and the light data of the lesion area, as well as the distance between the two edges and the size of the overlapping area, and color re-adjustment is performed to obtain multi-level atlas shared data.

[0104] Furthermore, step S1 further includes remotely operating the ultrasound device to acquire images, wherein the image acquisition module is an ultrasound probe and an ultrasound scanner, wherein the ultrasound probe is used to generate and receive ultrasound signals, and the ultrasound scanner is used to control signal amplification, filtering, and digital image reconstruction;

[0105] Step S2 also includes evaluating the quality of the ultrasound image and triggering a re-acquisition instruction when the quality is lower than a predetermined standard. The ultrasound image acquisition module also includes a remote control unit for remotely operating the ultrasound device to acquire images.

[0106] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A WEB-based ultrasonic image intelligent acquisition system, characterized in that: include: An image acquisition module is used to obtain multiple sets of image data and screen the multiple sets of image data to obtain qualified image data and unqualified image data; Analysis module, used to analyze qualified image data to obtain feature data , analyze the unqualified image data to obtain feature data ; The first processing module is used to process the feature data With feature data Perform recombinant contrast processing to obtain dark data of the lesion area and bright color data of the lesion area ; The second processing module is used to process the dark data of the lesion area The size of is assigned to obtain the primary lesion data; according to the bright color data of the lesion area The size of is assigned to obtain the secondary lesion data; According to the dark data of the lesion area and bright color data of the lesion area The distance between the two edges is assigned to obtain the third-level lesion data; according to the dark data of the lesion area , bright color data of lesion area The area with overlapping edges was assigned values ​​to obtain the fourth-level lesion data; An output module is used to perform color re-adjustment processing on lesion data at different levels to obtain multi-level atlas shared data; For feature data With feature data Perform recombinant contrast processing to obtain dark data of the lesion area Bright color data of lesion area ,include: Confirming the ultrasound image The collection of dark pixels is set as , preset the threshold of dark pixels , when the gray value of the pixel is less than the threshold of the dark pixel When , the pixel is considered as a dark pixel and added to the set In the ultrasound image, The total number of pixels is set as , according to the set of dark pixels , determine the number of dark pixels and set it as , the number of dark pixels Divide by the total number of pixels , get the dark data of the lesion area ; Confirming the ultrasound image The set of medium and bright pixels is set as , preset the threshold of bright pixels , when the gray value of the pixel is greater than the threshold of the bright pixel , the pixel is considered a bright pixel and added to the set In the ultrasound image, The total number of pixels is set as , according to the set of bright pixels , determine the number of bright pixels and set it , the number of bright pixels Divide by the total number of pixels , get the bright color data of the lesion area .

2. The WEB-based ultrasonic image intelligent acquisition system according to claim 1, characterized in that: Analyze qualified image data to obtain feature data ,include: For each set of image data in multiple sets of images, Processing to obtain feature data ,in, Represents the contrast of ultrasound images. The calculation formula is: ； in, Represents an ultrasound image, Represents ultrasound image In Location The pixel value of Represents ultrasound image The average pixel value, Represents ultrasound image The total number of pixels in and Represents ultrasound images In Location The horizontal and vertical gradients at , , Represents the weight coefficient.

3. The WEB-based ultrasonic image intelligent acquisition system according to claim 2, characterized in that: According to the dark data of the lesion area , bright color data of lesion area The distance between the two edges is assigned to obtain the third-level lesion data, including: Determine each pixel in the dark pixel set D And calculate the edge pixels , edge pixels The calculation formula is: ; Determine the bright pixel set Each pixel in And calculate the edge pixels , edge pixels The calculation formula is: ; in, and Represents pixel points The horizontal and vertical gradient values, and Represents pixel The horizontal gradient and vertical gradient values, and Represents pixel , The multi-level gradient processing value of Location The calculation formula is: ； in, represents the Gaussian kernel, represents the convolution operation, Represents an ultrasound image, Indicates the zoom ratio, Represents the arc direction angle of the pixel under the horizontal gradient and vertical gradient, which is obtained by dividing the horizontal gradient by the vertical gradient. Represents the statistical processing of the histogram function, Indicates contrast enhancement processing, by obtaining pixel points With pixels The pixel values ​​of the two edges are used to calculate the distance between the two edges. , and Divide by the number of pairs of dark pixels and light pixels to get the average distance , the number of pairings is the total number of groups of total distance, and the average distance obtained under different numbers of groups is Each set of ultrasound images is obtained by assigning values ​​according to the numerical values. The data of grade 3 lesions.

4. The WEB-based ultrasonic image intelligent acquisition system according to claim 3, characterized in that: According to ultrasound images Medium dark pixel collection The data is at the preset dark pixel threshold Based on , which is used to further subdivide the dark areas at different levels, where the pixel brightness is The following settings are for the first level dark area 1. Pixel brightness Above and in The following settings are for the secondary dark area 2. Pixel brightness The above settings are for the third-level dark area 3. For each subdivided dark area, extract the corresponding dark feature vector , The calculation formula is: ； in, Indicates the number of pixels in dark areas of different levels, For the coordinates The gray value of the enhanced image at Indicates The average gray value of the dark area at different levels is calculated based on the dark feature vectors of the pixels in the dark areas at different levels. Arrange them according to the numerical values, and assign values ​​according to the relationship between the numerical values ​​to obtain the primary lesion data.

5. The WEB-based ultrasonic image intelligent acquisition system according to claim 4, characterized in that: According to the size of the primary lesion data, the calculation formula is adjusted according to the color system. Processing, and combining pseudo color technology to generate different color areas, setting multiple readjustment thresholds Used to subdivide the classified data under different rescaling treatments; in The value size is The following settings are for the green edge area. The value size is Above and in The following settings are the yellow edge area. The value size is The above settings are for the red edge area, and the color system is adjusted by the calculation formula for: ； in, Indicates the number of pixels in dark areas of different levels, , These are color re-adjustment parameters. Represents the average grayscale value of all pixels in set D, which is used to reflect the overall brightness level of the pixel group currently being processed.

6. The WEB-based ultrasonic image intelligent acquisition system according to claim 5, characterized in that: According to the size of the primary lesion data, the calculation formula is adjusted according to the color system. Processing is performed and different color areas are generated in combination with pseudo color technology. The pseudo color technology includes a data acquisition unit and a processing unit, wherein the data acquisition unit further includes a signal conversion module and a scanning module, the signal conversion module is responsible for converting analog signals into digital signals, and the scanning module is used to scan image data of different dark areas. The processing unit further includes an image reconstruction module and a WEB sharing module, the image reconstruction module is used to perform B-mode imaging processing on digital signals, and the WEB sharing module is used to transmit the acquired digital signals to the WEB platform for remote synchronous processing.

7. A WEB-based ultrasonic image intelligent acquisition method, characterized in that: The method is applied to a WEB-based ultrasonic image intelligent acquisition system according to any one of claims 1 to 6, comprising the following steps: S1: acquiring image data through an image acquisition module, and screening the image data to obtain qualified image data and unqualified image data; S2: Analyze qualified image data to obtain feature data ; Used to analyze unqualified image data to obtain feature data ; S3: By analyzing the feature data With feature data Perform recombinant contrast processing to obtain dark data of the lesion area , bright color data of lesion area ; S4: Multi-level lesion data is obtained by assigning values ​​to the size of the dark data and the light data of the lesion area, as well as the distance between the two edges and the size of the overlapping area, and color re-adjustment is performed to obtain multi-level atlas shared data.

8. The WEB-based ultrasonic image intelligent acquisition method according to claim 7, characterized in that: Step S1 also includes remotely operating an ultrasound device to acquire images, wherein the image acquisition module is an ultrasound probe and an ultrasound scanner, wherein the ultrasound probe is used to generate and receive ultrasound signals, and the ultrasound scanner is used to control signal amplification, filtering, and digital image reconstruction; Step S2 also includes evaluating the quality of the ultrasound image and triggering a re-acquisition instruction when the quality is lower than a predetermined standard. The image acquisition module also includes a remote control unit for remotely operating the ultrasound device to acquire images.

Images