CT (Computed Tomography) image clear processing method, system and equipment for pulmonary infectious diseases

Abstract

The invention relates to the technical field of image processing, and provides a CT image clear processing method, system and equipment for pulmonary infectious diseases, and the method comprises the steps: screening out the edge pixel points of all pulmonary blood vessels according to an authenticity evaluation coefficient, and obtaining the edge pixel points of the pulmonary blood vessels; according to the gray level distribution condition of the pixel points in each lung blood vessel edge area and the authenticity evaluation coefficient, obtaining a frosted shadow feature value of each lung blood vessel edge area; obtaining a lung attrition lesion area; according to a difference condition of gray values of pixel points between the lung attrition lesion area and the surrounding lung attrition lesion area, obtaining a window width adjustment amplitude of the lung image of the patient; and performing window width adjustment on the lung image of the patient according to the window width adjustment amplitude to obtain the adjusted lung image of the patient. According to the invention, more detailed and clearer position and form of the lung lesion area can be provided through the adjusted lung image of the patient.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a method, system and device for clear processing of CT images of pulmonary infectious diseases. Background Art

[0002] Pulmonary infectious diseases, such as mycoplasma pneumonia and tuberculosis, have diverse clinical manifestations and develop rapidly. Although traditional X-ray examinations can initially determine whether there are abnormalities in the lungs, they are limited in detail display and resolution. CT imaging technology, especially CT images that have been clearly processed, can provide more detailed and accurate information about the lung structure, helping doctors to more accurately determine the location, morphology and range of lesions. Mycoplasma pneumonia is a common lung infection in children, especially in school-age children and young people. However, in the lung images of patients with Mycoplasma pneumoniae infection taken by CT equipment, the infected part appears blurred, which may cause misjudgment by doctors during the observation process, resulting in inaccurate judgment results. Summary of the invention

[0003] The present invention provides a method, system and device for clear processing of CT images of pulmonary infectious diseases, so as to solve the problem that the infected area appears blurred in the lung images of patients infected with Mycoplasma pneumoniae taken by CT equipment. The technical solutions adopted are as follows: The present invention proposes a method for processing CT images of pulmonary infectious diseases, which comprises the following steps: Obtaining images of the patient's lungs; Obtain edge pixel points of pulmonary blood vessels in the patient's lung image; obtain the authenticity evaluation coefficient of each edge pixel point of the pulmonary blood vessel and its corresponding edge pixel point according to the position distribution of the edge pixel point of each pulmonary blood vessel; screen out the edge pixel points of all pulmonary blood vessels according to the authenticity evaluation coefficient to obtain the edge pixel points of the pulmonary blood vessels; According to the edge pixel points of the pulmonary blood vessels, all pulmonary blood vessel edge areas are obtained; according to the grayscale distribution of the pixel points in each pulmonary blood vessel edge area and the authenticity evaluation coefficient, the ground-glass shadow characteristic value of each pulmonary blood vessel edge area is obtained; according to the ground-glass shadow characteristic value, all pulmonary blood vessel edge areas are screened to obtain the pulmonary ground-glass lesion area; according to the difference in the grayscale value of the pixel points between the pulmonary ground-glass lesion area and the surrounding pulmonary ground-glass lesion areas, the window width adjustment range of the patient's lung image is obtained; The window width of the patient's lung image is adjusted according to the window width adjustment amplitude to obtain an adjusted patient's lung image.

[0004] Preferably, the obtaining of the authenticity evaluation coefficient of each edge pixel point of the pulmonary blood vessel and its corresponding edge pixel point according to the position distribution of the edge pixel point of each pulmonary blood vessel comprises the following specific methods: For The edge pixel points of the first pulmonary blood vessel are obtained, and the edge pixel points of all other pulmonary blood vessels are obtained. The Euclidean distance between the edge pixels of the first pulmonary blood vessels; the edge pixels of the two pulmonary blood vessels with the smallest Euclidean distance are both recorded as The edge pixel point on the same side as the edge pixel point of the pulmonary blood vessel; The straight line connecting the two edge points on the same side of the edge pixel point of the pulmonary blood vessel is recorded as The straight line along the same side of the edge pixel point of the pulmonary blood vessel; The edge pixel points of the pulmonary blood vessels are perpendicular to the edge straight line on the same side and recorded as The vertical edge straight line of the edge pixel points of the first pulmonary blood vessel is obtained; the edge pixel points of all pulmonary blood vessels on the vertical edge straight line are obtained. The Euclidean distance between the edge pixels of the pulmonary vessels is calculated, and the edge pixel of the pulmonary vessel with the smallest Euclidean distance is recorded as the first The corresponding edge pixel points of the edge pixel points of the pulmonary blood vessels; The first The average gray value of all pixels on the straight line connecting the edge pixel of the pulmonary blood vessel and its corresponding edge pixel is recorded as The grayscale mean of the corresponding pixels at the edge of the pulmonary blood vessels; The Euclidean distance between the edge pixel of a pulmonary vessel and its corresponding edge pixel is recorded as The corresponding edge distance of the edge pixel points of the pulmonary blood vessels; The grayscale mean of the corresponding pixels between the edge pixels of the first pulmonary blood vessels is The normalized value of the ratio of the corresponding edge distances of the edge pixels of the pulmonary blood vessels is used as the first The authenticity evaluation coefficient of the edge pixel of a pulmonary blood vessel and its corresponding edge pixel.

[0005] Preferably, the step of screening out edge pixels of all pulmonary blood vessels according to the authenticity evaluation coefficient to obtain edge pixels of pulmonary blood vessels includes the following specific methods: Preset a threshold coefficient parameter , The authenticity evaluation coefficient of the edge pixel of the pulmonary blood vessel and its corresponding edge pixel is greater than or equal to the threshold coefficient parameter , will The edge pixel points of the pulmonary blood vessels and their corresponding edge pixel points are all recorded as edge pixel points of the pulmonary blood vessels.

[0006] Preferably, the step of obtaining all pulmonary blood vessel edge regions based on edge pixel points of the pulmonary blood vessels includes the following specific methods: Preset a window parameter , all the edge pixels of the pulmonary blood vessels constitute the edge of the pulmonary blood vessels, and are evenly divided into several pixels of size For any edge segment, the edge segment is taken as the size of A window is constructed based on the window edge of the pulmonary vascular region, and the formed area is recorded as the pulmonary vascular edge area; the pulmonary vascular edge area will not be located in the pulmonary vascular region.

[0007] Preferably, the method of obtaining the ground-glass shadow characteristic value of each pulmonary blood vessel edge region according to the grayscale distribution of the pixels in each pulmonary blood vessel edge region and the authenticity evaluation coefficient includes the following specific methods: According to the grayscale distribution of the pixels in each pulmonary blood vessel edge area, the ground-glass shadow characteristic factor of each pulmonary blood vessel edge area is obtained; The first The average of the authenticity evaluation coefficients of all edge pixels of the pulmonary blood vessels and their corresponding edge pixels in the edge area of ​​the pulmonary blood vessels is recorded as the first evaluation mean; The normalized value of the product of the ground-glass shadow characteristic factor of the edge area of ​​the pulmonary vessels and the first evaluation mean is used as the first The ground-glass opacity characteristic values ​​of the edge areas of the pulmonary vessels.

[0008] Preferably, the method of obtaining the ground-glass opacity characteristic factor of each pulmonary blood vessel edge region according to the grayscale distribution of the pixel points in each pulmonary blood vessel edge region includes the following specific methods: For Any pixel point in the edge area of ​​the pulmonary blood vessels, if the gray value of the pixel point is greater than or equal to the gray value parameter , record the pixel point as a feature pixel point; in the edge area of ​​the pulmonary blood vessels, obtain the Euclidean distance between the feature pixel point and all other feature pixel points, and record the feature pixel point with the smallest Euclidean distance as the target feature pixel point of the feature pixel point; record the Euclidean distance between the feature pixel point and its target feature pixel point as the feature distance value of the feature pixel point; The first The mean grayscale value of all pixels in the edge area of ​​the first pulmonary blood vessels is The product of the number of all characteristic pixels in the edge area of ​​the pulmonary blood vessels is recorded as the first product; The average of the characteristic distance values ​​of all characteristic pixels in the edge area of ​​the pulmonary blood vessels is recorded as the first distance mean; the ratio between the first product and the first distance mean is taken as the first distance mean. The ground-glass opacity characteristic factor of the edge area of ​​pulmonary vessels.

[0009] Preferably, the method of screening all pulmonary vascular edge regions according to the ground-glass shadow feature value to obtain the pulmonary ground-glass lesion region includes the following specific methods: Preset a threshold feature parameter , The ground-glass shadow feature value of the edge area of ​​the pulmonary blood vessels is greater than the threshold feature parameter , will The edge area of ​​the pulmonary blood vessels was recorded as the pulmonary ground-glass lesion area.

[0010] Preferably, the window width adjustment amplitude of the patient's lung image is obtained according to the difference in grayscale values ​​of pixels between the pulmonary ground glass lesion area and the surrounding pulmonary ground glass lesion areas, and the specific method includes: Preset a neighborhood parameter , for the The ground-glass lesion area of ​​the lung was obtained. The Euclidean distance between the centroid of the ground-glass lesion area and the centroid of all other ground-glass lesion areas of the lung is taken as the minimum Euclidean distance. Each area of ​​pulmonary ground glass lesions was used as the first Comparison of ground-glass lesion areas in the lungs; For Any one of the ground-glass lesion areas of the lung is compared with the ground-glass lesion area of ​​the lung. The absolute value of the difference between the grayscale value mean of all pixels in the first ground glass lesion area and the grayscale value mean of all pixels in the contrast ground glass lesion area is recorded as the grayscale difference value of the contrast ground glass lesion area. The reciprocal of the mean of all grayscale difference values ​​of the ground-glass lesion area of ​​the lung is recorded as the first reciprocal; the The product of the ground-glass shadow characteristic value of the ground-glass lesion area of ​​the lung and the first inverse is taken as the first Window width adjustment factor for ground-glass lesion areas in the lungs; The normalized value of the mean of the window width adjustment factors of all ground-glass lesion areas of the lung is used as the window width adjustment amplitude of the patient's lung image.

[0011] The present invention also proposes a CT image clear processing system for pulmonary infectious diseases, which includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps of the above method are implemented.

[0012] The present invention also proposes a CT image clear processing device for pulmonary infectious diseases, the device comprising: A patient lung image acquisition module, a window width adjustment amplitude acquisition module and an image window width adjustment module. The patient lung image acquisition module is used to acquire the patient lung image. The window width adjustment amplitude acquisition module implements the steps of the above method by calling a computer program to obtain the window width adjustment amplitude of the patient lung image. The image window width adjustment module adjusts the window width of the patient lung image based on the window width adjustment amplitude to obtain the adjusted patient lung image.

[0013] The beneficial effects of the present invention are as follows: the present invention obtains all pulmonary blood vessel edge areas according to the edge pixel points of the pulmonary blood vessels; obtains the ground-glass shadow characteristic value of each pulmonary blood vessel edge area according to the grayscale distribution of the pixel points in each pulmonary blood vessel edge area and the authenticity evaluation coefficient; screens all pulmonary blood vessel edge areas according to the ground-glass shadow characteristic value to obtain the pulmonary ground-glass lesion area; obtains the window width adjustment amplitude of the patient's lung image according to the difference in the grayscale value of the pixel points between the pulmonary ground-glass lesion area and the surrounding pulmonary ground-glass lesion areas; adjusts the window width of the patient's lung image according to the window width adjustment amplitude to obtain the adjusted patient's lung image; and the adjusted patient's lung image can provide a more detailed and clearer position and morphology of the lung lesion area. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0015] Figure 1 The present invention is a flowchart of the steps of the method for clear processing of CT images of pulmonary infectious diseases; Figure 2 The figure is a flow chart showing the characteristic relationship of the method for clear processing of CT images of pulmonary infectious diseases according to the present invention. DETAILED DESCRIPTION

[0016] 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 creative work are within the scope of protection of the present invention.

[0017] See also Figure 1 , which shows a flow chart of a method for clear processing of CT images of pulmonary infectious diseases provided by an embodiment of the present invention, the method comprising the following steps: Step S001, obtaining a patient's lung image.

[0018] The purpose of this embodiment is to determine the area where the ground-glass shadow is located by analyzing the specific information shown in the patient's lung CT image, so as to adjust the window width of the image, enhance the difference between the ground-glass shadow and the surrounding lung tissue in the image, and make the patient's lung CT image clearer.

[0019] Specifically, a CT machine is used to obtain a lung CT image of the patient, and the lung CT image of the patient is subjected to median filtering denoising and grayscale operations to obtain a grayscale image of the lung CT of the patient; the grayscale image of the lung CT of the patient is segmented using semantic segmentation technology to obtain a lung image of the patient.

[0020] Among them, median filtering, grayscale operation and semantic segmentation technology are existing technologies, and will not be described in detail in this embodiment.

[0021] At this point, an image of the patient's lungs is obtained.

[0022] Step S002, obtaining edge pixel points of the pulmonary blood vessels in the patient's lung image; obtaining the authenticity evaluation coefficient of each edge pixel point of the pulmonary blood vessel and its corresponding edge pixel point according to the position distribution of the edge pixel points of each pulmonary blood vessel; screening the edge pixel points of all pulmonary blood vessels according to the authenticity evaluation coefficient to obtain the edge pixel points of the pulmonary blood vessels.

[0023] It should be noted that when children suffer from bronchopneumonia, it is generally lobular pneumonia, which is generally accompanied by multiple flake-like increased density shadows, called "ground-glass shadows", which appear as localized or diffuse ground-glass opaque shadows in the lungs, while the internal blood vessels and bronchi are still visible; and ground-glass shadows may appear as cloudy, patchy or nodular highly dense shadows, and are usually distributed along the bronchial vascular bundles; because the lungs are filled with gas, the density is the lowest and it appears black in the image, the higher the density, the whiter the color, and the spine has the largest density and the whitest color; based on the above-mentioned characteristic manifestations, analysis is performed to determine the position of the ground-glass shadows on the patient's lung image; but before obtaining the position of the ground-glass shadows, it is necessary to obtain the pulmonary vascular area and its edge pixels.

[0024] Preferably, in one embodiment of the present invention, since ground-glass shadows are usually distributed along vascular bundles, the approximate position of the ground-glass shadows on the patient's lung image can be obtained by determining the position of the intrapulmonary blood vessels, and the intrapulmonary blood vessels are denser than the lungs and appear white in the image; therefore, the Otsu threshold segmentation algorithm is used to perform threshold segmentation on the patient's lung image to obtain the pulmonary vascular region in the patient's lung image; the Canny edge detection algorithm is used to perform edge detection on the pulmonary vascular region to obtain the edge pixel points of the pulmonary blood vessels.

[0025] Among them, the Otsu threshold segmentation algorithm and the Canny edge detection algorithm are existing technologies, and will not be described in detail in this embodiment.

[0026] It should be noted that due to the sensitivity of the edge detection algorithm to image edge details and noise, some edge pixels are blurred, so it is impossible to determine the authenticity of the edge pixels obtained by the edge detection algorithm; therefore, it is necessary to determine whether the edge pixels are true edge pixels; since the distribution of edge pixels of the pulmonary blood vessels on the pulmonary blood vessels has certain characteristic performances, that is, the edge pixels on the same side of the pulmonary blood vessels will have corresponding edge pixels on the other side, and the distance between the one-to-one corresponding edge pixels is close, and all pixels on the straight line connecting the two appear white in the threshold segmentation result image, that is, the grayscale value is high.

[0027] Preferably, in one embodiment of the present invention, according to the position distribution of the edge pixel points of each pulmonary blood vessel, the authenticity evaluation coefficient of the edge pixel point of each pulmonary blood vessel and its corresponding edge pixel point is obtained, and the specific method includes: For The edge pixel points of the first pulmonary blood vessel are obtained, and the edge pixel points of all other pulmonary blood vessels are obtained. The Euclidean distance between the edge pixels of the first pulmonary blood vessels; the edge pixels of the two pulmonary blood vessels with the smallest Euclidean distance are both recorded as The edge pixel point on the same side as the edge pixel point of the pulmonary blood vessel; The straight line connecting the two edge points on the same side of the edge pixel point of the pulmonary blood vessel is recorded as The straight line along the same side of the edge pixel point of the pulmonary blood vessel; The edge pixel points of the pulmonary blood vessels are perpendicular to the edge straight line on the same side and recorded as The vertical edge straight line of the edge pixel points of the first pulmonary blood vessel is obtained; the edge pixel points of all pulmonary blood vessels on the vertical edge straight line are obtained. The Euclidean distance between the edge pixels of the pulmonary vessels is calculated, and the edge pixel of the pulmonary vessel with the smallest Euclidean distance is recorded as the first The corresponding edge pixel points of the edge pixel points of the pulmonary blood vessels; The first The average gray value of all pixels on the straight line connecting the edge pixel of the pulmonary blood vessel and its corresponding edge pixel is recorded as The grayscale mean of the corresponding pixels at the edge of the pulmonary blood vessels; The Euclidean distance between the edge pixel of a pulmonary vessel and its corresponding edge pixel is recorded as The corresponding edge distance of the edge pixel points of the pulmonary blood vessels; The grayscale mean of the corresponding pixels between the edge pixels of the first pulmonary blood vessels is The normalized value of the ratio of the corresponding edge distances of the edge pixels of the pulmonary blood vessels is used as the first The authenticity evaluation coefficient of the edge pixel of each pulmonary blood vessel and its corresponding edge pixel; The specific formula is: In the formula, Indicates The authenticity evaluation coefficient of the edge pixel of each pulmonary blood vessel and its corresponding edge pixel; Indicates All pixels on the straight line connecting the edge pixel of a pulmonary blood vessel and its corresponding edge pixel; Indicates The first pixel on the line connecting the edge pixel of the pulmonary blood vessel and its corresponding edge pixel The gray value of each pixel; Indicates The Euclidean distance between the edge pixel of a pulmonary vessel and its corresponding edge pixel; represents the linear normalization function.

[0028] It should be noted that since pulmonary blood vessels are generally thin, the distance between the edge points of pulmonary blood vessels and their corresponding edge pixels should be closer, and the density of blood vessels is greater than the density of lung gas, and they themselves appear white. Therefore, after threshold segmentation, the pixels on the straight line connecting the two also appear white; and when the distance between the corresponding edge pixels is closer, the grayscale value of the pixel between the two is larger, and the authenticity evaluation coefficient of the pixel belonging to the edge of the pulmonary blood vessels is greater.

[0029] Preferably, in one embodiment of the present invention, the edge pixels of all pulmonary blood vessels are screened out according to the authenticity evaluation coefficient, and the specific method includes: Preset a threshold coefficient parameter , wherein this embodiment is based on This example is described as an example, and this embodiment is not specifically limited. Depends on the specific implementation situation; Jordi The authenticity evaluation coefficient of the edge pixel of the pulmonary blood vessel and its corresponding edge pixel is greater than or equal to the threshold coefficient parameter , will The edge pixel points of the pulmonary blood vessels and their corresponding edge pixel points are all recorded as edge pixel points of the pulmonary blood vessels.

[0030] At this point, the edge pixels of all pulmonary blood vessels are obtained.

[0031] Step S003, based on the edge pixel points of the pulmonary blood vessels, all pulmonary blood vessel edge areas are obtained; based on the grayscale distribution of the pixels in each pulmonary blood vessel edge area and the authenticity evaluation coefficient, the ground-glass shadow characteristic value of each pulmonary blood vessel edge area is obtained; based on the ground-glass shadow characteristic value, all pulmonary blood vessel edge areas are screened to obtain the pulmonary ground-glass lesion area; based on the difference in the grayscale value of the pixels between the pulmonary ground-glass lesion area and the surrounding pulmonary ground-glass lesion areas, the window width adjustment range of the patient's lung image is obtained.

[0032] It should be noted that ground-glass shadows are usually distributed along the edges of pulmonary blood vessels. By analyzing the performance of surrounding pixels along the pulmonary blood vessels in the patient's lung images, it can be determined whether ground-glass shadow features exist. Ground-glass shadow features are due to their relatively large regional density, and they appear white, and are cloudy, patchy or nodular, and are relatively dense. Based on the above characteristics, the approximate location of the ground-glass shadow can be analyzed, so that the ground-glass shadow can be clarified to improve the visibility and clarity of ground-glass shadow lesions.

[0033] Preferably, in one embodiment of the present invention, the specific method for obtaining the edge regions of all pulmonary blood vessels according to the edge pixels of the pulmonary blood vessels is: Preset a window parameter , wherein this embodiment is based on This example is described as an example, and this embodiment is not specifically limited. Depends on the specific implementation situation; All the edge pixels of the pulmonary blood vessels constitute the edge of the pulmonary blood vessels and are evenly divided into several For any edge segment, the edge segment is taken as the size of A window is constructed based on the window edge of the pulmonary vascular region, and the formed area is recorded as the pulmonary vascular edge area; the pulmonary vascular edge area will not be located in the pulmonary vascular region.

[0034] Preferably, in one embodiment of the present invention, according to the grayscale distribution of pixels in each pulmonary blood vessel edge region and the authenticity evaluation coefficient, the specific method for obtaining the ground-glass shadow characteristic value of each pulmonary blood vessel edge region is: Preset a grayscale parameter , wherein this embodiment is based on This example is described as an example, and this embodiment is not specifically limited. Depends on the specific implementation situation; For Any pixel point in the edge area of ​​the pulmonary blood vessels, if the gray value of the pixel point is greater than or equal to the gray value parameter , record the pixel point as a feature pixel point; in the edge area of ​​the pulmonary blood vessels, obtain the Euclidean distance between the feature pixel point and all other feature pixel points, and record the feature pixel point with the smallest Euclidean distance as the target feature pixel point of the feature pixel point; record the Euclidean distance between the feature pixel point and its target feature pixel point as the feature distance value of the feature pixel point; The first The mean grayscale value of all pixels in the edge area of ​​the first pulmonary blood vessels is The product of the number of all characteristic pixels in the edge area of ​​the pulmonary blood vessels is recorded as the first product; The average of the characteristic distance values ​​of all characteristic pixels in the edge area of ​​the pulmonary blood vessels is recorded as the first distance mean; the ratio between the first product and the first distance mean is taken as the first distance mean. The ground-glass opacity characteristic factor of the edge area of ​​the pulmonary vessels; The specific formula is: In the formula, Indicates The ground-glass opacity characteristic factor of the edge area of ​​the pulmonary vessels; Indicates The mean gray value of all pixels in the edge area of ​​pulmonary blood vessels; Indicates The number of all characteristic pixels in the edge area of ​​pulmonary blood vessels; Indicates The average of the characteristic distance values ​​of all characteristic pixels in the edge area of ​​the pulmonary blood vessels.

[0035] It should be noted that The larger the mean gray value of all pixels in the edge area of ​​the first pulmonary blood vessels, the larger the mean gray value of all pixels in the edge area of ​​the first pulmonary blood vessels. The more the number of all characteristic pixels in the edge area of ​​the pulmonary blood vessels is, the smaller the mean of the characteristic distance values ​​of all characteristic pixels is, indicating that The more the characteristics of the edge area of ​​the lung blood vessels conform to the ground-glass shadow, the denser the distribution is, and the color appears white, that is, the larger the gray value.

[0036] Preferably, in one embodiment of the present invention, when the ground-glass opacity factor of the edge region of the pulmonary blood vessels is larger and the authenticity evaluation coefficient of the edge pixel point of the pulmonary blood vessels and its corresponding edge pixel point is higher, it indicates that the reliability of the existence of ground-glass opacity in the edge region of the pulmonary blood vessels is higher; then The specific method for calculating the ground-glass shadow characteristic value of the edge area of ​​the pulmonary blood vessels is: The first The average of the authenticity evaluation coefficients of all edge pixels of the pulmonary blood vessels and their corresponding edge pixels in the edge area of ​​the pulmonary blood vessels is recorded as the first evaluation mean; The normalized value of the product of the ground-glass shadow characteristic factor of the edge area of ​​the pulmonary vessels and the first evaluation mean is used as the first The ground-glass opacity characteristic value of the edge area of ​​the pulmonary blood vessels; The specific formula is: In the formula, Indicates The ground-glass opacity characteristic value of the edge area of ​​the pulmonary blood vessels; Indicates The ground-glass opacity characteristic factor of the edge area of ​​the pulmonary vessels; Indicates The average of the authenticity evaluation coefficients of all edge pixels of the pulmonary blood vessels and their corresponding edge pixels in the edge area of ​​the pulmonary blood vessels; represents the linear normalization function.

[0037] Preferably, in one embodiment of the present invention, all pulmonary vascular edge regions are screened according to ground-glass shadow feature values ​​to obtain pulmonary ground-glass lesion regions, including the following specific methods: Preset a threshold feature parameter , wherein this embodiment is based on This example is described as an example, and this embodiment is not specifically limited. Depends on the specific implementation situation; Jordi The ground-glass shadow feature value of the edge area of ​​the pulmonary blood vessels is greater than the threshold feature parameter , will The edge area of ​​the pulmonary blood vessels was recorded as the pulmonary ground-glass lesion area.

[0038] It should be noted that when the contrast between the ground glass lesion area of ​​the lung and the surrounding ground glass lesion area is large, the window width needs to be adjusted slightly to appropriately increase the difference between the ground glass lesion area of ​​the lung and the surrounding lung tissue; when the contrast between the ground glass lesion area of ​​the lung and the surrounding ground glass lesion area is small, the window width needs to be adjusted to a greater extent to strongly increase the difference between the ground glass lesion area of ​​the lung and the surrounding lung tissue.

[0039] Preferably, in one embodiment of the present invention, according to the difference in grayscale values ​​of pixels between the ground-glass lesion area of ​​the lung and the surrounding ground-glass lesion area of ​​the lung, a specific method for obtaining the window width adjustment range of the patient's lung image is: Preset a neighborhood parameter , wherein this embodiment is based on This example is described as an example, and this embodiment is not specifically limited. Depends on the specific implementation situation; For The ground-glass lesion area of ​​the lung was obtained The Euclidean distance between the centroid of the ground-glass lesion area and the centroid of all other ground-glass lesion areas of the lung is taken as the minimum Euclidean distance. Each area of ​​pulmonary ground glass lesions was used as the first Comparison of ground-glass lesion areas in the lungs; For Any one of the ground-glass lesion areas of the lung is compared with the ground-glass lesion area of ​​the lung. The absolute value of the difference between the grayscale value mean of all pixels in the first ground glass lesion area and the grayscale value mean of all pixels in the contrast ground glass lesion area is recorded as the grayscale difference value of the contrast ground glass lesion area. The reciprocal of the mean of all grayscale difference values ​​of the ground-glass lesion area of ​​the lung is recorded as the first reciprocal; the The product of the ground-glass shadow characteristic value of the ground-glass lesion area of ​​the lung and the first inverse is taken as the first Window width adjustment factor for ground-glass lesion areas in the lungs; The normalized value of the mean of the window width adjustment factors of all ground-glass lesion areas of the lung is used as the window width adjustment amplitude of the patient's lung image; The specific formula is: In the formula, Indicates the window width adjustment amplitude of the patient's lung image; Indicates all areas of ground-glass lesions in the lungs; Represents the preset neighborhood parameters; Indicates The mean gray value of all pixels in the ground-glass lesion area of ​​the lung; Indicates The first The mean gray value of all pixels in the ground-glass lesion area of ​​the contrast lung; Indicates The ground-glass opacity characteristic value of the edge area of ​​the pulmonary blood vessels; Indicates taking the absolute value; represents the linear normalization function.

[0040] It should be noted that Indicates The window width adjustment factor for the ground-glass lesion area of ​​the lung.

[0041] At this point, the window width adjustment range of the patient's lung image is obtained.

[0042] Step S004: adjusting the window width of the patient's lung image according to the window width adjustment amplitude to obtain an adjusted patient's lung image.

[0043] It should be noted that the window width of the patient's lung image is adjusted to increase the difference between the ground-glass shadow and the surrounding lung tissue, making the ground-glass shadow features in the patient's lung image clearer.

[0044] Preferably, in one embodiment of the present invention, the window width of the patient's lung image is adjusted according to the window width adjustment amplitude, and the specific method for obtaining the adjusted patient's lung image is: The difference between 1 and the window width adjustment amplitude of the patient's lung image is recorded as the first difference; the product of the first difference and the original window width of the patient's lung image is used as the adjusted window width of the patient's lung image; the patient's lung image is adjusted according to the adjusted window width of the patient's lung image to obtain the adjusted patient's lung image.

[0045] Among them, adjusting the patient's lung image according to the adjustment window width of the patient's lung image is a prior art, and this embodiment will not be described in detail here.

[0046] See also Figure 2, which shows a characteristic relationship flow chart of a CT image clear processing method for pulmonary infectious diseases; At this point, this embodiment is completed.

[0047] Another embodiment of the present invention provides a CT image clarity processing system for pulmonary infectious diseases, the system comprising a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, the above method steps S001 to S004 are implemented.

[0048] Another embodiment of the present invention provides a CT image clear processing device for pulmonary infectious diseases, the device comprising: A patient lung image acquisition module, a window width adjustment amplitude acquisition module and an image window width adjustment module. The patient lung image acquisition module is used to acquire the patient lung image. The window width adjustment amplitude acquisition module implements steps S001 to S004 of the above method by calling a computer program to obtain the window width adjustment amplitude of the patient lung image. The image window width adjustment module adjusts the window width of the patient lung image based on the window width adjustment amplitude to obtain the adjusted patient lung image.

[0049] 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 method for processing CT images of pulmonary infectious diseases, characterized in that: The method comprises the following steps: Acquire an image of the patient's lungs; Obtain edge pixel points of pulmonary blood vessels in the patient's lung image; obtain the authenticity evaluation coefficient of each edge pixel point of the pulmonary blood vessel and its corresponding edge pixel point according to the position distribution of the edge pixel point of each pulmonary blood vessel; screen out the edge pixel points of all pulmonary blood vessels according to the authenticity evaluation coefficient to obtain the edge pixel points of the pulmonary blood vessels; According to the edge pixel points of the pulmonary blood vessels, all pulmonary blood vessel edge areas are obtained; according to the grayscale distribution of the pixel points in each pulmonary blood vessel edge area and the authenticity evaluation coefficient, the ground-glass shadow characteristic value of each pulmonary blood vessel edge area is obtained; according to the ground-glass shadow characteristic value, all pulmonary blood vessel edge areas are screened to obtain the pulmonary ground-glass lesion area; according to the difference in the grayscale value of the pixel points between the pulmonary ground-glass lesion area and the surrounding pulmonary ground-glass lesion areas, the window width adjustment range of the patient's lung image is obtained; Adjusting the window width of the patient's lung image according to the window width adjustment amplitude to obtain an adjusted patient's lung image; The method of obtaining the authenticity evaluation coefficient of each edge pixel point of the pulmonary blood vessel and its corresponding edge pixel point according to the position distribution of the edge pixel point of each pulmonary blood vessel includes the following specific methods: For The edge pixel points of the first pulmonary blood vessel are obtained, and the edge pixel points of all other pulmonary blood vessels are obtained. The Euclidean distance between the edge pixels of the first pulmonary blood vessels; the edge pixels of the two pulmonary blood vessels with the smallest Euclidean distance are both recorded as The edge pixel point on the same side as the edge pixel point of the pulmonary blood vessel; The straight line connecting the two edge points on the same side of the edge pixel point of the pulmonary blood vessel is recorded as The straight line along the same side of the edge pixel point of the pulmonary blood vessel; The edge pixel points of the pulmonary blood vessels are perpendicular to the edge straight line on the same side and recorded as The vertical edge straight line of the edge pixel points of the first pulmonary blood vessel is obtained; the edge pixel points of all pulmonary blood vessels on the vertical edge straight line are obtained. The Euclidean distance between the edge pixels of the pulmonary vessels is calculated, and the edge pixel of the pulmonary vessel with the smallest Euclidean distance is recorded as the first The corresponding edge pixel points of the edge pixel points of the pulmonary blood vessels; The first The average gray value of all pixels on the straight line connecting the edge pixel of the pulmonary blood vessel and its corresponding edge pixel is recorded as The grayscale mean of the corresponding pixels at the edge of the pulmonary blood vessels; The Euclidean distance between the edge pixel of a pulmonary vessel and its corresponding edge pixel is recorded as The corresponding edge distance of the edge pixel points of the pulmonary blood vessels; The grayscale mean of the corresponding pixels between the edge pixels of the first pulmonary blood vessels is The normalized value of the ratio of the corresponding edge distances of the edge pixels of the pulmonary blood vessels is used as the first The authenticity evaluation coefficient of the edge pixel of each pulmonary blood vessel and its corresponding edge pixel; The method of obtaining the ground-glass shadow characteristic value of each pulmonary blood vessel edge region according to the grayscale distribution of the pixel points in each pulmonary blood vessel edge region and the authenticity evaluation coefficient includes: According to the grayscale distribution of the pixels in each pulmonary blood vessel edge area, the ground-glass shadow characteristic factor of each pulmonary blood vessel edge area is obtained; The first The average of the authenticity evaluation coefficients of all edge pixels of the pulmonary blood vessels and their corresponding edge pixels in the edge area of ​​the pulmonary blood vessels is recorded as the first evaluation mean; The normalized value of the product of the ground-glass shadow characteristic factor of the edge area of ​​the pulmonary vessels and the first evaluation mean is used as the first The ground-glass opacity characteristic value of the edge area of ​​the pulmonary blood vessels; The method of obtaining the ground-glass shadow characteristic factor of each pulmonary blood vessel edge region according to the grayscale distribution of the pixel points in each pulmonary blood vessel edge region includes: For Any pixel point in the edge area of ​​the pulmonary blood vessels, if the gray value of the pixel point is greater than or equal to the gray value parameter , record the pixel point as a feature pixel point; in the edge area of ​​the pulmonary blood vessels, obtain the Euclidean distance between the feature pixel point and all other feature pixel points, and record the feature pixel point with the smallest Euclidean distance as the target feature pixel point of the feature pixel point; record the Euclidean distance between the feature pixel point and its target feature pixel point as the feature distance value of the feature pixel point; The first The mean grayscale value of all pixels in the edge area of ​​the first pulmonary blood vessels is The product of the number of all characteristic pixels in the edge area of ​​the pulmonary blood vessels is recorded as the first product; The average of the characteristic distance values ​​of all characteristic pixels in the edge area of ​​the pulmonary blood vessels is recorded as the first distance mean; the ratio between the first product and the first distance mean is taken as the first distance mean. The ground-glass opacity characteristic factor of the edge area of ​​pulmonary vessels.

2. The method for clear processing of CT images of pulmonary infectious diseases according to claim 1, characterized in that: The method of screening out edge pixels of all pulmonary blood vessels according to the authenticity evaluation coefficient to obtain edge pixels of pulmonary blood vessels includes: Preset a threshold coefficient parameter , The authenticity evaluation coefficient of the edge pixel of the pulmonary blood vessel and its corresponding edge pixel is greater than or equal to the threshold coefficient parameter , will The edge pixel points of the pulmonary blood vessels and their corresponding edge pixel points are all recorded as edge pixel points of the pulmonary blood vessels.

3. The method for clear processing of CT images of pulmonary infectious diseases according to claim 1, characterized in that: The specific method of obtaining all pulmonary blood vessel edge regions according to the edge pixel points of the pulmonary blood vessels is as follows: Preset a window parameter , all the edge pixels of the pulmonary blood vessels constitute the edge of the pulmonary blood vessels, and are evenly divided into several pixels of size For any edge segment, the edge segment is taken as the size of A window is constructed based on the window edge of the pulmonary vascular region, and the formed area is recorded as the pulmonary vascular edge area; the pulmonary vascular edge area will not be located in the pulmonary vascular region.

4. The method for clear processing of CT images of pulmonary infectious diseases according to claim 1, characterized in that: The method of screening all pulmonary vascular edge regions according to the ground-glass shadow feature value to obtain the pulmonary ground-glass lesion region includes: Preset a threshold feature parameter , The ground-glass shadow feature value of the edge area of ​​the pulmonary blood vessels is greater than the threshold feature parameter , will The edge area of ​​the pulmonary blood vessels was recorded as the pulmonary ground-glass lesion area.

5. The method for clear processing of CT images of pulmonary infectious diseases according to claim 1, characterized in that: The specific method of obtaining the window width adjustment range of the patient's lung image according to the difference in grayscale values ​​of pixels between the ground-glass lesion area of ​​the lung and the surrounding ground-glass lesion areas of the lung includes: Preset a neighborhood parameter , for the The ground-glass lesion area of ​​the lung was obtained The Euclidean distance between the centroid of the ground-glass lesion area and the centroid of all other ground-glass lesion areas of the lung is taken as the minimum Euclidean distance. Each area of ​​pulmonary ground glass lesions was used as the first Comparison of ground-glass lesion areas in the lungs; For Any one of the ground-glass lesion areas of the lung is compared with the ground-glass lesion area of ​​the lung. The absolute value of the difference between the grayscale value mean of all pixels in the first ground glass lesion area and the grayscale value mean of all pixels in the contrast ground glass lesion area is recorded as the grayscale difference value of the contrast ground glass lesion area. The reciprocal of the mean of all grayscale difference values ​​of the ground-glass lesion area of ​​the lung is recorded as the first reciprocal; the The product of the ground-glass shadow characteristic value of the ground-glass lesion area of ​​the lung and the first inverse is taken as the first Window width adjustment factor for ground-glass lesion areas in the lungs; The normalized value of the mean of the window width adjustment factors of all ground-glass lesion areas of the lung is used as the window width adjustment amplitude of the patient's lung image.

6. A CT image clear processing system for pulmonary infectious diseases, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the CT image clear processing method for pulmonary infectious diseases as described in any one of claims 1-5 are implemented.

7. CT image clear processing equipment for pulmonary infectious diseases, the equipment comprising: A patient lung image acquisition module, a window width adjustment amplitude acquisition module and an image window width adjustment module. The patient lung image acquisition module is used to acquire the patient lung image. The window width adjustment amplitude acquisition module implements the steps of the CT image clear processing method for pulmonary infectious diseases as described in any one of claims 1 to 5 by calling a computer program to obtain the window width adjustment amplitude of the patient lung image. The image window width adjustment module adjusts the window width of the patient lung image based on the window width adjustment amplitude to obtain the adjusted patient lung image.

Images