Intelligent thyroid ultrasound image analysis system

Abstract

The invention relates to the technical field of image data processing, in particular to a thyroid ultrasound image intelligent analysis system. The system obtains an ultrasonic image of a thyroid gland, and obtains a calcification edge feature value according to the distribution position and gray level distribution of edge pixel points in a local area of pixel points in the ultrasonic image of the thyroid gland; according to the complexity and the discrete condition of the gray value of the pixel point in each preset direction in the local area of the pixel point, the confusion index of each preset direction is obtained, the organization boundary ambiguity is obtained in combination with the calcified edge feature value, and then the gain coefficient in the ACE algorithm is adjusted to obtain the improved gain coefficient, so that the accuracy and the robustness of the algorithm are improved. And carrying out image enhancement on the ultrasonic image of the thyroid gland by using the improved ACE algorithm. The gain coefficient in the ACE algorithm is improved based on the fuzzy degree of the tissue boundary of the ultrasonic image of the thyroid gland, and the contrast ratio and the image quality of the ultrasonic image are effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of image data processing, and in particular to a thyroid ultrasound image intelligent analysis system. Background Art

[0002] Thyroid nodules are a common complication that greatly endangers people's lives. Thyroid ultrasound imaging is an important means of judging thyroid nodules, but the ultrasound images formed by ultrasound imaging have the disadvantage of weak visual performance, that is, the contrast of ultrasound images is low and the clarity is poor. Therefore, it is necessary to enhance the contrast of thyroid ultrasound images and improve the quality of ultrasound images.

[0003] The existing technology uses the traditional adaptive contrast enhancement (ACE) algorithm to enhance the ultrasound image of the thyroid gland. However, the local grayscale value difference in the blurred tissue boundary area is large, which makes the gain coefficient small, and the enhancement effect on the blurred tissue boundary area of ​​the thyroid nodule in the ultrasound image is poor, resulting in poor enhancement effect of the ultrasound image of the thyroid gland. Summary of the invention

[0004] In order to solve the technical problem that the local gray value difference in the area with blurred tissue boundaries is large, resulting in poor enhancement effect of ultrasound images, the purpose of the present invention is to provide a thyroid ultrasound image intelligent analysis system, and the technical solution adopted is as follows: The present invention proposes a thyroid ultrasound image intelligent analysis system, the system comprising: An image data acquisition module, used for acquiring an ultrasonic image of the thyroid gland; A calcification feature analysis module is used to obtain a calcification edge feature value of each pixel in the ultrasound image according to the distribution position and grayscale distribution of edge pixels in a preset window of each pixel in the ultrasound image; A grayscale chaos analysis module is used to obtain a chaos index of each pixel in the ultrasound image in each preset direction according to the complexity and discreteness of the grayscale value of the pixel in each preset direction within a preset window of each pixel in the ultrasound image; A tissue boundary fuzzy analysis module, used to obtain the tissue boundary fuzziness of each pixel point in the ultrasound image by combining the chaos index of each pixel point in all preset directions and the calcification edge feature value; The ultrasonic image enhancement module is used to improve the gain coefficient in the ACE algorithm based on the grayscale distribution in a preset window of each pixel point in the ultrasonic image and the fuzziness of the tissue boundary, and use the obtained improved gain coefficient to enhance the ultrasonic image of the thyroid gland.

[0005] Furthermore, the step of obtaining the calcification edge feature value of each pixel in the ultrasound image according to the distribution position and grayscale distribution of edge pixels in a preset window of each pixel in the ultrasound image includes: For each pixel point in the ultrasound image, two adjacent edge pixel points in a preset window of the pixel point are connected to form a straight line segment as an edge straight line segment, and all edge straight line segments in the preset window of the pixel point are obtained; the method for determining the two adjacent edge pixel points is: if one edge pixel point is located in a preset neighborhood of the other edge pixel point, the two edge pixel points are adjacent; The angle between each edge straight line segment and the preset reference straight line in the preset window of the pixel point is used as the edge angle of each edge straight line segment; the ratio of the edge angle to the preset angle is used as the angle ratio of each edge straight line segment; According to the difference between the angle ratios of the edge straight line segments in the preset window of each pixel point in the ultrasound image and the grayscale distribution, the calcification edge feature value of each pixel point in the ultrasound image is obtained.

[0006] Furthermore, the calculation formula of the calcified edge feature value of the pixel point is as follows: ; ; In the formula, is the calcification edge feature value of the i-th pixel in the ultrasound image; n is the number of pixels in the preset window of the i-th pixel in the ultrasound image; is the gray value of the xth pixel in the preset window of the ith pixel in the ultrasound image; is the gray value of the i-th pixel in the ultrasound image; is the edge slope difference of the i-th pixel in the ultrasound image; is the number of edge straight line segments in the preset window of the i-th pixel point in the ultrasound image; is the angle ratio of the uth edge straight line segment in the preset window of the i-th pixel point in the ultrasound image; is the angle ratio of the u-1th edge straight line segment in the preset window of the i-th pixel point in the ultrasound image; e is a natural constant; is the normalization function; is the absolute value function.

[0007] Furthermore, the method for obtaining the confusion index includes: For each pixel in the ultrasound image, the grayscale value of the pixel in the preset window of the pixel is sampled from front to back according to each preset direction to obtain a sampling sequence of each preset direction of the pixel; Arrange the absolute values ​​of the differences between each element and the value of the next element of the element in the sampling sequence of each preset direction of the pixel point in order to obtain a differential grayscale sequence of each preset direction of the pixel point; Substitute the probability of occurrence of the value of each element in the differential grayscale sequence of each preset direction of the pixel point into the information entropy formula to obtain the chaos degree of the differential grayscale sequence of each preset direction of the pixel point; The mean value of the values ​​of all elements in the differential grayscale sequence of each preset direction of the pixel point is obtained as the average grayscale value of the differential grayscale sequence of the corresponding preset direction of the pixel point; the absolute value of the difference between the value of each element in the differential grayscale sequence of each preset direction of the pixel point and the average grayscale value is used as the grayscale difference of the corresponding element; the cumulative sum of the grayscale difference values ​​of all elements in the differential grayscale sequence of each preset direction of the pixel point is used as the comprehensive grayscale difference feature value of the differential grayscale sequence of each preset direction of the pixel point; According to the chaos degree and the comprehensive grayscale difference characteristic value of the differential grayscale sequence of each preset direction of the pixel point, the chaos index of each preset direction of the pixel point is obtained; the chaos degree and the comprehensive grayscale difference characteristic value are both positively correlated with the chaos index.

[0008] Furthermore, the method for obtaining the fuzziness of tissue boundaries includes: The cumulative sum of the chaos indexes of all preset directions of each pixel in the ultrasound image is used as the comprehensive chaos index of each pixel; According to the calcification edge characteristic value and the comprehensive chaos index of each pixel in the ultrasound image, the tissue boundary fuzziness of each pixel in the ultrasound image is obtained; the calcification edge characteristic value and the comprehensive chaos index are both positively correlated with the tissue boundary fuzziness.

[0009] Furthermore, the grayscale distribution in a preset window of each pixel in the ultrasonic image and the fuzziness of the tissue boundary are used to improve the gain coefficient in the ACE algorithm, and the ultrasonic image of the thyroid is enhanced using the obtained improved gain coefficient, including: The ACE algorithm is used to enhance the ultrasound image and obtain the gain coefficient of each pixel in the ultrasound image; The product of the gain coefficient of each pixel point in the ultrasound image and the fuzziness of the tissue boundary is used as the improved gain coefficient of each pixel point in the ultrasound image; The ultrasound images are enhanced using the ACE algorithm based on the improved gain coefficient of each pixel.

[0010] The present invention has the following beneficial effects: In the embodiment of the present invention, based on the common calcification phenomenon in the ultrasonic image of the thyroid gland, the present invention analyzes the color distribution characteristics and irregular characteristics of the calcification edge according to the distribution position and grayscale distribution of the edge pixel points in the local area of ​​the pixel point, and obtains the calcification edge characteristic value of the pixel point; and analyzes the degree of blurring of the local tissue boundary of the thyroid gland. Since the blurring has multi-directionality, the complexity and discreteness of the grayscale value of the pixel point in the local area of ​​the pixel point in multiple preset directions are used to analyze the blurring of the tissue boundary in each preset direction in the local area of ​​the pixel point, and obtain the chaos index of the pixel point in each preset direction; and comprehensively analyzes the calcification edge characteristic value of the pixel point and the chaos index of each preset direction to obtain the tissue boundary blurring of the pixel point, improves the gain coefficient of the traditional ACE algorithm, obtains the improved gain coefficient of the pixel point, and enhances the ultrasonic image of the thyroid gland using the ACE algorithm based on the improved gain coefficient; avoids the gain coefficient of the smooth area to be large, adjusts it with a smaller tissue boundary blur, and can effectively enhance the tissue boundary blurring area to a greater extent, so that the tissue structure in the enhanced ultrasonic image of the thyroid gland is clearer, and improves the contrast of the ultrasonic image. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0012] Figure 1 The system block diagram of a thyroid ultrasound image intelligent analysis system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0013] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation method, structure, features and effects of a thyroid ultrasound image intelligent analysis system proposed by the present invention are described in detail below in combination 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.

[0014] 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.

[0015] The specific scheme of the thyroid ultrasound image intelligent analysis system provided by the present invention is described in detail below with reference to the accompanying drawings.

[0016] See also Figure 1 , which shows a system block diagram of a thyroid ultrasound image intelligent analysis system provided by an embodiment of the present invention, the system includes: an image data acquisition module 101, a calcification feature analysis module 102, a grayscale chaos analysis module 103, a tissue boundary fuzzy analysis module 104, and an ultrasound image enhancement module 105.

[0017] The image data acquisition module 101 is used to acquire an ultrasonic image of the thyroid gland.

[0018] Specifically, the purpose of intelligently analyzing the ultrasonic image of the thyroid gland is to enhance the ultrasonic image of the thyroid gland and improve the quality of the ultrasonic image of the thyroid gland to assist doctors in making judgments on the patient's condition. The embodiment of the present invention uses an ultrasonic probe device to perform ultrasonic exploration of the patient's thyroid gland and forms an initial ultrasonic image of the thyroid gland through an ultrasonic imaging device. It should be noted that the initial ultrasonic image of the thyroid gland formed by the ultrasonic imaging device is a grayscale image. Therefore, the initial ultrasonic image does not need to be grayscaled and can be directly analyzed.

[0019] In order to avoid geometric deformation of the initial ultrasound image due to external factors, the perspective transformation technology is used to correct the initial ultrasound image to obtain a corrected ultrasound image. In order to eliminate the influence of noise on subsequent analysis and retain the boundary information to the greatest extent, the corrected ultrasound image needs to be denoised.

[0020] In the embodiment of the present invention, bilateral filtering is used to perform denoising on the corrected ultrasonic image. In other embodiments of the present invention, other filtering algorithms may be selected to perform denoising on the corrected ultrasonic image, such as median filtering, mean filtering and Gaussian filtering.

[0021] The Canny edge detection operator is used to perform edge detection on the corrected ultrasound image, and the detected edge pixels are marked in the corrected ultrasound image, so that the ultrasound image of the thyroid gland can be obtained based on the analysis of the distribution characteristics of the edge pixels in the local area of ​​the pixels. The Canny edge detection operator is a well-known technology for those skilled in the art and will not be described in detail here.

[0022] At this point, an ultrasound image of the thyroid gland is obtained.

[0023] The calcification feature analysis module 102 is used to obtain the calcification edge feature value of each pixel in the ultrasound image according to the distribution position and grayscale distribution of edge pixels in a preset window of each pixel in the ultrasound image.

[0024] Specifically, when the traditional ACE algorithm is used to enhance the ultrasound image of the thyroid gland, the gain coefficient of the high-frequency area in the ultrasound image is small, resulting in a weaker enhancement effect at the edge of the thyroid gland; the smoother the smooth area, the greater the gain coefficient, which is more likely to produce a ringing effect. Therefore, the contrast of the ultrasound image of the thyroid gland is low and the clarity is poor. Among them, the ACE algorithm is a well-known technology for those skilled in the art and will not be described in detail here.

[0025] Enhancing the edge details in the ultrasound image of the thyroid gland can make the tissue boundaries in the ultrasound image of the thyroid gland clearer, making it easier for medical students to learn about the structure of thyroid nodules. The traditional ACE algorithm focuses on enhancing all areas with low contrast and does not distinguish between enhanced areas, so the traditional ACE algorithm needs to be improved.

[0026] Calcification is an extremely important clinical feature of thyroid disease. Calcification is divided into microcalcification and coarse calcification. Microcalcification is more likely to produce strong echoes in the form of dots or clusters in the local area of ​​the thyroid ultrasound image, while coarse calcification is more likely to produce relatively large calcification areas in the thyroid gland, and is more likely to form irregular plaque-like calcification foci in the local area of ​​the thyroid gland. The edges of the above calcification areas have less contrast and clarity in the ultrasound image, so their contrast needs to be enhanced.

[0027] There are many edge pixels in the local area of ​​pixels at the edge of the calcified area and their distribution is relatively messy. At the same time, the difference in grayscale values ​​in the local area of ​​pixels is large. Combining the above two factors, the analysis of pixels at the edge of calcification is more accurate.

[0028] Preferably, the specific method for obtaining the calcification edge characteristic value is: for each pixel point in the ultrasound image, two adjacent edge pixel points in a preset window of the pixel point are connected, and the straight line segment formed is used as the edge straight line segment, and all the edge straight line segments in the preset window of the pixel point are obtained; the method for judging two adjacent edge pixel points is: if one of the edge pixel points is located in the preset neighborhood of the other edge pixel point, the two edge pixel points are adjacent; the angle between each edge straight line segment in the preset window of the pixel point and the preset reference straight line is used as the edge angle of each edge straight line segment; the ratio of the edge angle to the preset angle is used as the angle ratio of each edge straight line segment; according to the difference between the angle ratios of the edge straight line segments in the preset window of each pixel point in the ultrasound image and the grayscale distribution, the calcification edge characteristic value of each pixel point in the ultrasound image is obtained.

[0029] As an example, when calcification occurs in the thyroid gland, there are many edge pixels in the local area of ​​the pixel points at the edge of the calcified area, and the distribution is relatively messy. The present invention presents the situation that the pixel point is located at the edge of the calcified area by the inclination of the straight line formed by adjacent edge pixels in the local area of ​​the pixel point.

[0030] A preset window is constructed with each pixel in the ultrasound image as the center. Taking the i-th pixel in the ultrasound image as an example, the grayscale value difference between the i-th pixel and the pixels in its preset window, as well as the difference in the degree of inclination between the straight lines formed by adjacent edge pixels in the preset window of the i-th pixel are analyzed to obtain the calcification edge feature value of the i-th pixel In the embodiment of the present invention, the size of the preset window of pixel points is an empirical value. , implementers can adjust according to actual conditions.

[0031] (1) Analyze the grayscale distribution of pixels within a preset window.

[0032] The grayscale difference of the edge area formed by calcification is large. When the difference between the grayscale value of a pixel and its local area is greater, the possibility that the pixel is located at the edge of the calcified area is greater. Pixel point x is a pixel point in the preset window of the i-th pixel point. Calculate the grayscale value of the i-th pixel point The gray value of each pixel in its preset window The absolute value of the difference , and sum up the absolute values ​​of these differences to get ,in is the number of pixels in the preset window of the i-th pixel. The larger the value, the greater the possibility that the i-th pixel is located at the edge of the calcified area.

[0033] (2) Analyze the position distribution of adjacent edge pixels within the preset window of pixels.

[0034] Assume that edge pixel points a and b are both edge pixel points within the preset neighborhood of the i-th pixel point in the ultrasound image. If edge pixel point b is located within the preset neighborhood of edge pixel point a, the preset neighborhood of the edge pixel point in the embodiment of the present invention takes the empirical value of the eight neighborhoods, and the implementer can set it according to the specific situation. That is, if edge pixel point b is located within the eight neighborhoods of edge pixel point a, then edge pixel points a and b are considered to be adjacent.

[0035] The straight line segment formed by connecting edge pixel point a and edge pixel point b is taken as the edge straight line segment. According to the above method, all edge straight line segments in the preset window of the i-th pixel point are obtained. For each edge straight line segment, the angle between the edge straight line segment and the preset reference straight line L is taken as the edge angle of the edge straight line segment. For the convenience of subsequent analysis, the edge angle of the edge straight line segment is normalized using the preset angle to obtain the angle ratio of the edge straight line segment. The range of the angle between two straight lines is , then the preset angle in the embodiment of the present invention is an empirical value .

[0036] In the embodiment of the present invention, the preset reference straight line is set as a horizontal line. In other embodiments, a vertical line can be used as the preset reference straight line. The implementer can also set a straight line in other directions as the preset reference straight line according to actual conditions.

[0037] (3) Thyroid calcification will produce strong echoes in the form of dots or clusters in the local area of ​​the ultrasound image, or form irregular plaque-like calcification foci in the local area. When a pixel is located at the edge of the calcified area, the inclination of the edge straight line segment in the preset area of ​​the pixel is irregular, and the difference between the angle ratios of the edge straight line segments in the preset window of the pixel is large. When the number of edge straight line segments in the preset neighborhood of the pixel is greater than 1, the absolute value of the difference in the angle ratios between the edge straight line segments is calculated and accumulated. The larger the accumulated result, the greater the possibility that the pixel is located at the edge of the calcified area; when the number of edge straight line segments in the preset neighborhood of the pixel is less than or equal to 1, the absolute value of the difference in the angle ratios of the edge straight line segments cannot be calculated.

[0038] According to the difference between the angle ratios of the edge straight line segments in the preset window of each pixel point in the ultrasound image and the grayscale distribution, the calcification edge feature value of each pixel point in the ultrasound image is obtained. The calculation formula of the calcification edge feature value is as follows: ; ; In the formula, is the calcification edge feature value of the i-th pixel in the ultrasound image; n is the number of pixels in the preset window of the i-th pixel in the ultrasound image; is the gray value of the xth pixel in the preset window of the ith pixel in the ultrasound image; is the gray value of the i-th pixel in the ultrasound image; is the edge slope difference of the i-th pixel in the ultrasound image; is the number of edge straight line segments in the preset window of the i-th pixel point in the ultrasound image; is the angle ratio of the uth edge straight line segment in the preset window of the i-th pixel point in the ultrasound image; is the angle ratio of the u-1th edge straight line segment in the preset window of the i-th pixel point in the ultrasound image; e is a natural constant; is the normalization function; is the absolute value function.

[0039] It should be noted that the grayscale difference in the edge area of ​​calcification is relatively large. When the value is larger, it means that the difference between the grayscale values ​​of the ith pixel in the ultrasound image and the pixel x in its preset neighborhood is greater, and the possibility that the ith pixel is located at the calcified edge of the thyroid gland is greater. The larger the size, the more calcified the thyroid gland will be. When the value is larger, it means that the edge pixel points in the preset window of the i-th pixel point are distributed chaotically, making the inclination of the edge straight line segment more irregular. The i-th pixel point is more likely to be located at the calcified edge of the thyroid gland, and the calcified edge feature value is The bigger.

[0040] At this point, the calcification edge feature value of each pixel in the ultrasound image of the thyroid gland is obtained.

[0041] The grayscale chaos analysis module 103 is used to obtain the chaos index of each pixel in the ultrasound image in each preset direction according to the complexity and discreteness of the grayscale value of each pixel in the ultrasound image in each preset direction within a preset window.

[0042] The nodules of thyroid cancer are generally highly invasive to tissue cells, which can easily lead to blurred tissue boundaries. The blurred phenomenon is mainly due to the chaotic local differences, which make it impossible to clearly identify the various tissue structures of the thyroid gland. Since the blurred direction is not unique, the present invention sets a plurality of preset directions, and performs fuzzy analysis on the grayscale values ​​of the pixels in each preset direction within the preset window of the pixels to obtain the chaos index of each preset direction; the complexity and discreteness of the grayscale values ​​of the pixels in each preset direction can reflect the chaos of the grayscale values ​​in each preset direction.

[0043] It should be noted that, in the embodiment of the present invention, the preset directions take empirical values ​​of horizontal and vertical directions.

[0044] Preferably, the method for obtaining the chaos index is as follows: for each pixel point in the ultrasonic image, the grayscale values ​​of the pixel points in the preset window of the pixel points are sampled from front to back according to each preset direction, and the sampling sequence of each preset direction of the pixel points is obtained; in the sampling sequence of each preset direction of the pixel points, the absolute values ​​of the difference between the values ​​of each element and the next element of the element are arranged in order, and the differential grayscale sequence of each preset direction of the pixel points is obtained; the probability of occurrence of the value of each element in the differential grayscale sequence of each preset direction of the pixel points is substituted into the information entropy formula, and the chaos degree of the differential grayscale sequence of each preset direction of the pixel points is obtained; the differential grayscale sequence of each preset direction of the pixel points is obtained. The mean of the numerical values ​​of all elements in the sequence is used as the average grayscale value of the differential grayscale sequence of the corresponding preset direction of the pixel point; the absolute value of the difference between the numerical value of each element in the differential grayscale sequence of each preset direction of the pixel point and the average grayscale value is used as the grayscale difference of the corresponding element; the cumulative sum of the grayscale difference values ​​of all elements in the differential grayscale sequence of each preset direction of the pixel point is used as the comprehensive grayscale difference eigenvalue of the differential grayscale sequence of each preset direction of the pixel point; according to the chaos degree and the comprehensive grayscale difference eigenvalue of the differential grayscale sequence of each preset direction of the pixel point, the chaos index of each preset direction of the pixel point is obtained; the chaos degree and the comprehensive grayscale difference eigenvalue are both positively correlated with the chaos index.

[0045] As an example, taking the ith pixel in the ultrasound image as an example for analysis, the gray value distribution of the pixel in the preset window of the ith pixel is , where q is the number of rows of pixels in the preset window, and p is the number of columns of pixels in the preset window. Since the size of the preset window is an empirical value, , then q=11, p=11; is the gray value of the pixel in the first row and first column in the preset window of the i-th pixel, is the gray value of the pixel in the 1st row and the pth column in the preset window of the i-th pixel, is the gray value of the pixel in the qth row and the first column in the preset window of the i-th pixel, is the grayscale value of the pixel at the qth row and the pth column in the preset window of the i-th pixel.

[0046] Taking the horizontal preset direction as an example for analysis, the horizontal sampling sequence of the i-th pixel point in the ultrasound image is ,in, is the gray value of the pixel in the 2nd row and 1st column in the preset window of the i-th pixel, is the gray value of the pixel in the 2nd row and the pth column in the preset window of the i-th pixel. The change of gray value in the horizontal sampling sequence mainly reflects the horizontal blurring of the tissue boundary. In order to more clearly reflect the horizontal gray chaos phenomenon in the preset window of the i-th pixel and avoid the influence of the original gray value on the analysis, the horizontal sampling sequence is The absolute values ​​of the differences between each element and the next element in the sequence are arranged to obtain a horizontal differential grayscale sequence. Among them, the difference gray sequence The first element in , differential grayscale sequence The second element , and so on, the differential grayscale sequence The kth element in is the last element .in, is the grayscale value of the pixel in the first row and second column in the preset window of the i-th pixel, is the grayscale value of the pixel in the first row and third column in the preset window of the i-th pixel, is the gray value of the pixel at the qth row and the p-1th column in the preset window of the i-th pixel; is the absolute value function. is the number of elements in the differential grayscale sequence.

[0047] (1) Analyze the chaos of the differential grayscale sequence of each pixel in each preset direction.

[0048] The more complex and chaotic the values ​​of the elements in the horizontal differential grayscale sequence are, the more serious the horizontal blurring of the tissue boundary in the preset window of the pixel point is. Therefore, the probability of occurrence of the value of each element in the horizontal differential grayscale sequence is substituted into the information entropy formula to obtain the chaos degree of the horizontal differential grayscale sequence. The calculation formula of the chaos degree is as follows: ; In the formula, is the disorder degree of the differential grayscale sequence of the t-th preset direction of the ith pixel in the ultrasound image; g is the type of the value of the element in the differential grayscale sequence of the t-th preset direction of the ith pixel in the ultrasound image, that is, the number of gray levels in the differential grayscale sequence of the t-th preset direction; is the probability of occurrence of the value of the vth element in the differential grayscale sequence of the tth preset direction of the ith pixel in the ultrasound image, and ln is a logarithmic function with the natural constant e as the base.

[0049] It should be noted that when g is larger, the values ​​of the elements in the differential grayscale sequence in the t-th preset direction are more chaotic, and the blurring of the tissue boundary in the t-th preset direction in the preset window of the pixel point is more serious. The bigger; when The closer it is to 0 or 1, the more regular the values ​​of the elements in the differential grayscale sequence in the t-th preset direction are, and the clearer the organizational boundary in the t-th preset direction in the preset window of the pixel point is. The smaller.

[0050] (2) Analyze the comprehensive grayscale value of the differential grayscale sequence of each pixel in each preset direction.

[0051] The degree of discreteness of the element values ​​in the differential grayscale sequence of each preset direction of the i-th pixel in the ultrasound image presents the blurring of the tissue boundary in the preset direction in the local area of ​​the pixel. The more serious the discreteness of the element values ​​in the differential grayscale sequence, the more serious the blurring of the tissue boundary in the preset window of the pixel. Calculate the mean value of the values ​​of all elements in the horizontal differential grayscale sequence of the i-th pixel as the average grayscale value of the horizontal differential grayscale sequence of the i-th pixel, obtain the absolute value of the difference between the value of each element in the differential grayscale sequence and the average grayscale value, and obtain the grayscale difference of each element; and add the grayscale differences of all elements in the differential grayscale sequence to obtain the comprehensive grayscale feature value of the horizontal differential grayscale sequence of the i-th pixel.

[0052] In other embodiments of the present invention, since both the standard deviation and the variance can reflect the discreteness of the data, the standard deviation or variance of the numerical values ​​of all elements in the differential grayscale sequence of each pixel point in each preset direction in the ultrasound image can be used as the comprehensive grayscale value of the differential grayscale sequence of the corresponding pixel point in each preset direction.

[0053] (3) Analyze the chaos index of each pixel in each preset direction.

[0054] Combined with the difference between the numerical values ​​of the elements in the differential grayscale sequence of each preset direction of the pixel point, and the degree of chaos, the chaos index of each preset direction of the pixel point is obtained. Since the greater the degree of chaos of the differential grayscale sequence of each preset direction of the i-th pixel point in the ultrasound image, the greater the comprehensive grayscale eigenvalue, the more serious the blurring of the tissue boundary in the preset window of the i-th pixel point, the chaos index of each preset direction of the i-th pixel point; that is, the degree of chaos and the comprehensive grayscale eigenvalue are both positively correlated with the chaos index. In an embodiment of the present invention, the product of the degree of chaos of the differential grayscale sequence of each preset direction of the pixel point and the comprehensive grayscale eigenvalue is used as the chaos index of each preset direction of the pixel point; in another embodiment of the present invention, the sum of the degree of chaos and the comprehensive grayscale eigenvalue can be used as the chaos index, so that the degree of chaos and the comprehensive grayscale eigenvalue are finally positively correlated with the chaos index. In other embodiments of the present invention, other basic mathematical operations can be selected to form a positive correlation, which is not limited here. The calculation formula of the chaos index is as follows: ; In the formula, is the chaos index of the ith pixel in the ultrasound image at the tth preset direction; is the disorder degree of the differential grayscale sequence of the ith pixel in the ultrasonic image in the tth preset direction; k is the number of elements in the differential grayscale sequence of the ith pixel in the ultrasonic image in the tth preset direction; is the value of the rth element in the differential grayscale sequence of the ith pixel in the tth preset direction in the ultrasound image; is the average grayscale value of the differential grayscale sequence of the i-th pixel in the ultrasound image in the t-th preset direction; is the grayscale difference of the rth element in the differential grayscale sequence of the i-th pixel in the t-th preset direction in the ultrasound image; is the comprehensive grayscale difference feature value of the differential grayscale sequence of the i-th pixel point in the ultrasound image in the t-th preset direction; is the absolute value function.

[0055] It should be noted that when When the value is larger, the value of the element in the differential grayscale sequence in the t-th preset direction is more chaotic, indicating that the blurring of the tissue boundary in the t-th preset direction in the preset window of the pixel point is more serious. The bigger; when The larger the value, the greater the discreteness of the value of the element in the differential grayscale sequence. The difference in the grayscale value in the preset direction of the preset window of the i-th pixel in the ultrasound image is large, the grayscale value is more chaotic, and the chaos index is The bigger.

[0056] According to the above method for obtaining the lateral chaos index of the i-th pixel point, the chaos index of each pixel point in each preset direction in the ultrasonic image of the thyroid gland is obtained.

[0057] The tissue boundary fuzzy analysis module 104 is used to obtain the tissue boundary fuzziness of each pixel point in the ultrasound image by combining the chaos index of each pixel point in all preset directions and the calcification edge feature value.

[0058] The calcification edge feature value presents the possibility that the pixel is located at the edge of the calcification area of ​​the thyroid gland. The confusion index represents the blurriness of the tissue boundary in the preset direction within the preset window of the pixel. The combined analysis of the above two indicators yields a higher accuracy of the tissue boundary blurriness.

[0059] Preferably, the specific method for obtaining the tissue boundary fuzziness is: taking the cumulative sum of the chaos indexes of all preset directions of each pixel point in the ultrasound image as the comprehensive chaos index of each pixel point; obtaining the tissue boundary fuzziness of each pixel point in the ultrasound image according to the calcification edge characteristic value and the comprehensive chaos index of each pixel point in the ultrasound image; the calcification edge characteristic value and the comprehensive chaos index are both positively correlated with the tissue boundary fuzziness.

[0060] The chaos index of all preset directions of each pixel is accumulated to obtain the comprehensive chaos index of the corresponding pixel. The chaos index of each preset direction of the pixel presents the blurring of the tissue boundary in the preset direction within the preset window of the pixel. The blurring of all preset directions is comprehensively analyzed, and the comprehensive chaos index of the pixel represents the overall blurring of the tissue boundary within the preset window of the pixel.

[0061] When the calcification edge characteristic value of the pixel point is larger and the comprehensive chaos index is larger, the overall blurring of the tissue boundary in the preset window of the pixel point is more serious, that is, the calcification edge characteristic value of the pixel point and the comprehensive chaos index are both positively correlated with the blurriness of the tissue boundary; in this embodiment of the present invention, the product of the calcification edge characteristic value and the comprehensive chaos index is normalized, and the normalized result is used as the blurriness of the tissue boundary; in another embodiment of the present invention, the result of normalizing the sum of the calcification edge characteristic value and the comprehensive chaos index can be used as the blurriness of the tissue boundary. In other embodiments of the present invention, other basic mathematical operations can be selected to form a positive correlation, which is not limited here.

[0062] The comprehensive chaos index of the pixel point and the calcification edge characteristic value are combined to obtain the tissue boundary fuzziness of the pixel point. The calculation formula of the tissue boundary fuzziness is as follows: ; In the formula, is the tissue boundary fuzziness of the i-th pixel in the ultrasound image; The calcification edge feature value of the i-th pixel in the ultrasound image; is the number of preset directions; is the comprehensive chaos index of the i-th pixel in the ultrasound image; is the normalization function.

[0063] It should be noted that when the calcification edge characteristic value The larger the value, the more likely it is that the i-th pixel in the ultrasound image is located at the edge of the calcified area, and the calcified tissue boundary is more blurred. The larger the chaos index When the value is larger, it means that the fuzzy boundary of the organization in the tth preset direction in the local area of ​​the pixel point is larger, making the comprehensive chaos index The larger the value is, the more serious the overall blurring of the tissue boundary within the preset window of the pixel point is, and the blurriness of the tissue boundary is The bigger.

[0064] The ultrasonic image enhancement module 105 is used to improve the gain coefficient in the ACE algorithm based on the grayscale distribution in a preset window of each pixel point in the ultrasonic image and the fuzziness of the tissue boundary, and enhance the ultrasonic image of the thyroid gland using the obtained improved gain coefficient.

[0065] The tissue boundary fuzziness presents the blurring of tissue boundaries in the local area of ​​a pixel point in the ultrasound image of the thyroid gland, and the pixel point is enhanced based on the tissue boundary fuzziness of the pixel point.

[0066] The ACE algorithm is used to enhance the ultrasound image to obtain the gain coefficient of each pixel in the ultrasound image; the product of the gain coefficient of each pixel in the ultrasound image and the fuzziness of the tissue boundary is used as the improved gain coefficient of each pixel in the ultrasound image.

[0067] The blurriness of tissue boundaries reflects the blurring phenomenon of local tissue boundaries to a certain extent. Ultrasonic images are grayscale images formed by ultrasonic detection, which are prone to local blurring and unclearness. Therefore, it is necessary to enhance the image of the unclear tissue boundary area to a greater extent. Therefore, the gain coefficient in the ACE algorithm is improved based on the blurriness of tissue boundaries to obtain the improved gain coefficient of each pixel. The calculation formula of the improved gain coefficient of each pixel in the ultrasound image is as follows: ; In the formula, is the improved gain coefficient of the i-th pixel in the ultrasound image; is the gain coefficient of the i-th pixel when using the traditional ACE algorithm to enhance the ultrasound image; is the tissue boundary fuzziness of the i-th pixel in the ultrasound image.

[0068] It should be noted that when organizational boundaries are blurred When the value is larger, the blurriness of the tissue boundary in the local area of ​​the pixel is greater. The gain coefficient is improved so that the improved gain coefficient of the i-th pixel point located at the edge of the calcified area of ​​the thyroid gland is larger. The greater the enhancement degree of the i-th pixel point by the ACE algorithm, the greater the contrast enhancement.

[0069] The ultrasound images are enhanced using the ACE algorithm based on the improved gain coefficient of each pixel.

[0070] According to the enhanced thyroid ultrasound image, the boundary of the thyroid tissue in the image is relatively clear, and the important characteristics of the thyroid nodules can be observed more clearly. Therefore, when teachers use the enhanced thyroid ultrasound image to teach medical students about thyroid nodules, students can observe thyroid nodules more clearly, which is helpful for medical students to learn about thyroid nodules.

[0071] At this point, the intelligent analysis system for thyroid ultrasound images is completed.

[0072] In summary, in an embodiment of the present invention, the present invention relates to the field of image data processing technology, and specifically to a thyroid ultrasound image intelligent analysis system. The system acquires an ultrasound image of the thyroid gland, and acquires a calcification edge characteristic value according to the distribution position and grayscale distribution of edge pixels in a local area of ​​pixels in the ultrasound image of the thyroid gland; obtains a confusion index for each preset direction according to the complexity and discreteness of the grayscale values ​​of pixels in each preset direction in the local area of ​​the pixels, and obtains the fuzziness of the tissue boundary in combination with the calcification edge characteristic value, and then adjusts the gain coefficient in the ACE algorithm to obtain an improved gain coefficient, so as to enhance the ultrasound image of the thyroid gland using the improved ACE algorithm. The present invention improves the gain coefficient in the ACE algorithm based on the fuzziness of the tissue boundary of the ultrasound image of the thyroid gland, effectively improving the contrast and image quality of the ultrasound image.

[0073] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0074] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0075] 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 thyroid ultrasound image intelligent analysis system, characterized in that: The system includes: An image data acquisition module, used for acquiring an ultrasonic image of the thyroid gland; A calcification feature analysis module is used to obtain a calcification edge feature value of each pixel in the ultrasound image according to the distribution position and grayscale distribution of edge pixels in a preset window of each pixel in the ultrasound image; A grayscale chaos analysis module is used to obtain a chaos index of each pixel in the ultrasound image in each preset direction according to the complexity and discreteness of the grayscale value of the pixel in each preset direction within a preset window of each pixel in the ultrasound image; A tissue boundary fuzzy analysis module, used to obtain the tissue boundary fuzziness of each pixel point in the ultrasound image by combining the chaos index of each pixel point in all preset directions and the calcification edge feature value; The ultrasonic image enhancement module is used to improve the gain coefficient in the ACE algorithm based on the grayscale distribution in a preset window of each pixel point in the ultrasonic image and the fuzziness of the tissue boundary, and use the obtained improved gain coefficient to enhance the ultrasonic image of the thyroid gland.

2. The thyroid ultrasound image intelligent analysis system according to claim 1, characterized in that: The step of obtaining the calcification edge feature value of each pixel in the ultrasound image according to the distribution position and grayscale distribution of edge pixels in a preset window of each pixel in the ultrasound image comprises: For each pixel point in the ultrasound image, two adjacent edge pixel points in a preset window of the pixel point are connected to form a straight line segment as an edge straight line segment, and all edge straight line segments in the preset window of the pixel point are obtained; the method for determining the two adjacent edge pixel points is: if one edge pixel point is located in a preset neighborhood of the other edge pixel point, the two edge pixel points are adjacent; The angle between each edge straight line segment and the preset reference straight line in the preset window of the pixel point is used as the edge angle of each edge straight line segment; the ratio of the edge angle to the preset angle is used as the angle ratio of each edge straight line segment; According to the difference between the angle ratios of the edge straight line segments in the preset window of each pixel point in the ultrasound image and the grayscale distribution, the calcification edge feature value of each pixel point in the ultrasound image is obtained.

3. The thyroid ultrasound image intelligent analysis system according to claim 2, characterized in that: The calculation formula of the calcified edge characteristic value of the pixel point is as follows: ； ； In the formula, is the calcification edge feature value of the i-th pixel in the ultrasound image; n is the number of pixels in the preset window of the i-th pixel in the ultrasound image; is the gray value of the xth pixel in the preset window of the ith pixel in the ultrasound image; is the gray value of the i-th pixel in the ultrasound image; is the edge slope difference of the i-th pixel in the ultrasound image; is the number of edge straight line segments in the preset window of the i-th pixel point in the ultrasound image; is the angle ratio of the uth edge straight line segment in the preset window of the i-th pixel point in the ultrasound image; is the angle ratio of the u-1th edge straight line segment in the preset window of the i-th pixel point in the ultrasound image; e is a natural constant; is the normalization function; is the absolute value function.

4. The thyroid ultrasound image intelligent analysis system according to claim 1, characterized in that: The method for obtaining the chaos index comprises: For each pixel in the ultrasound image, the grayscale value of the pixel in the preset window of the pixel is sampled from front to back according to each preset direction to obtain a sampling sequence of each preset direction of the pixel; Arrange the absolute values ​​of the differences between each element and the value of the next element of the element in the sampling sequence of each preset direction of the pixel point in order to obtain a differential grayscale sequence of each preset direction of the pixel point; Substitute the probability of occurrence of the value of each element in the differential grayscale sequence of each preset direction of the pixel point into the information entropy formula to obtain the chaos degree of the differential grayscale sequence of each preset direction of the pixel point; The mean value of the values ​​of all elements in the differential grayscale sequence of each preset direction of the pixel point is obtained as the average grayscale value of the differential grayscale sequence of the corresponding preset direction of the pixel point; the absolute value of the difference between the value of each element in the differential grayscale sequence of each preset direction of the pixel point and the average grayscale value is used as the grayscale difference of the corresponding element; the cumulative sum of the grayscale difference values ​​of all elements in the differential grayscale sequence of each preset direction of the pixel point is used as the comprehensive grayscale difference feature value of the differential grayscale sequence of each preset direction of the pixel point; According to the chaos degree and the comprehensive grayscale difference characteristic value of the differential grayscale sequence of each preset direction of the pixel point, the chaos index of each preset direction of the pixel point is obtained; the chaos degree and the comprehensive grayscale difference characteristic value are both positively correlated with the chaos index.

5. The thyroid ultrasound image intelligent analysis system according to claim 1, characterized in that: The method for obtaining the fuzziness of organizational boundaries comprises: The cumulative sum of the chaos indexes of all preset directions of each pixel in the ultrasound image is used as the comprehensive chaos index of each pixel; According to the calcification edge characteristic value and the comprehensive chaos index of each pixel in the ultrasound image, the tissue boundary fuzziness of each pixel in the ultrasound image is obtained; the calcification edge characteristic value and the comprehensive chaos index are both positively correlated with the tissue boundary fuzziness.

6. The thyroid ultrasound image intelligent analysis system according to claim 1, characterized in that: The method improves the gain coefficient in the ACE algorithm based on the grayscale distribution in the preset window of each pixel in the ultrasonic image and the fuzziness of the tissue boundary, and enhances the ultrasonic image of the thyroid gland using the obtained improved gain coefficient, including: The ACE algorithm is used to enhance the ultrasound image and obtain the gain coefficient of each pixel in the ultrasound image; The product of the gain coefficient of each pixel point in the ultrasound image and the fuzziness of the tissue boundary is used as the improved gain coefficient of each pixel point in the ultrasound image; The ultrasound images are enhanced using the ACE algorithm based on the improved gain coefficient of each pixel.

Images