Lung parameter measurement system and method based on lung CT cross section and medium

Abstract

The invention relates to the technical field of image processing, and discloses a lung parameter measurement system and method based on a lung CT cross section and a medium. The system comprises an image acquisition module used for acquiring a lung CT image; the left and right lung segmentation module is used for processing the lung CT image by using a deep learning model or an image processing algorithm to generate left and right lung segmentation masks; the bone segmentation module is used for processing the lung CT image by using a deep learning model or an image processing algorithm to generate a bone segmentation mask; the front-back diameter positioning module is used for positioning the front-back diameter of the front-back midline according to the left-right lung segmentation mask and the bone segmentation mask, and determining the front-back diameter of the left lung and the front-back diameter of the right lung; and the transverse diameter positioning module is used for positioning the thoracic transverse diameter according to the left and right lung segmentation masks, the skeleton segmentation masks and the front and back diameters of the front and back midlines, and determining the left lung transverse diameter and the right lung transverse diameter. The reliability of lung development parameter measurement is improved, and an accurate and consistent parameter measurement method is provided for children lung development research.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and particularly relates to a lung parameter measurement system, method and medium based on cross-sectional lung CT images. Background Art

[0002] Computed tomography (CT) is a technology for reconstructing three-dimensional medical images through digital geometric processing. In lung research, plain CT is often used for observing lung morphology due to its advantages such as low cost, controllable radiation dose, and fast imaging speed. However, existing research mainly focuses on the detection and segmentation of lung lesions, and there is less quantitative analysis of the geometric parameters of the lung cross-section (such as anteroposterior diameter and transverse diameter) in the study of the natural evolution law of children's lung development.

[0003] Traditional measurement methods usually rely on manual operations, which have problems such as strong subjectivity, large errors, low efficiency, lack of standardization, and insufficient automation, and are difficult to meet the requirements of efficient, accurate, and standardized measurement in the study of children's lung development. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to overcome the deficiencies in the prior art and provide a lung parameter measurement system, method and medium based on cross-sectional lung CT images.

[0005] The present invention provides the following technical solutions:

[0006] In a first aspect, an embodiment of the present disclosure provides a lung parameter measurement system based on cross-sectional lung CT images, the system includes:

[0007] An image acquisition module, configured to acquire lung CT images;

[0008] A left and right lung segmentation module, configured to process the lung CT images by using a first deep learning model or a first image processing algorithm to generate a left and right lung segmentation mask;

[0009] A bone segmentation module, configured to process the lung CT images by using a second deep learning model or a second image processing algorithm to generate a bone segmentation mask;

[0010] An anteroposterior diameter positioning module, configured to locate the endpoints of the anteroposterior diameter of the anteroposterior midline according to the left and right lung segmentation mask and the bone segmentation mask, and determine the left lung anteroposterior diameter and the right lung anteroposterior diameter according to the anteroposterior diameter of the anteroposterior midline;

[0011] A transverse diameter positioning module, configured to locate the endpoints of the thoracic transverse diameter according to the left and right lung segmentation mask, the bone segmentation mask and the endpoints of the anteroposterior diameter of the anteroposterior midline, and determine the left lung transverse diameter and the right lung transverse diameter according to the thoracic transverse diameter.

[0012] Optionally, the left and right lung segmentation module includes:

[0013] A left and right lung segmentation preprocessing unit for filtering the lung CT image to remove noise points in the lung CT image, obtaining a filtered lung CT image, and normalizing the filtered lung CT image to obtain a normalized lung CT image;

[0014] A left and right lung segmentation processing unit for processing the normalized lung CT image using the first deep learning model or the first image processing algorithm to generate an initial left and right lung segmentation mask;

[0015] A left and right lung segmentation post-processing unit for performing morphological operations on the initial left and right lung segmentation mask to remove noise points and isolated regions in the initial left and right lung segmentation mask, obtaining the left and right lung segmentation mask.

[0016] Optionally, the bone segmentation module includes:

[0017] A bone segmentation preprocessing unit for performing histogram equalization on the lung CT image to obtain an enhanced lung CT image;

[0018] A bone segmentation processing unit for processing the enhanced lung CT image using the second deep learning model or the second image processing algorithm to generate an initial bone segmentation mask;

[0019] A bone segmentation post-processing unit for performing morphological operations on the initial bone segmentation mask to remove noise points and isolated regions in the initial bone segmentation mask, obtaining the bone segmentation mask.

[0020] Optionally, the anteroposterior diameter positioning module includes:

[0021] An alternative slice determination unit for screening out slices that contain at least left lung pixels or right lung pixels in the left and right lung segmentation mask as candidate slices, and using slices within a preset range in the candidate slices as alternative slices;

[0022] An anteroposterior diameter determination unit for the anterior and posterior midlines to determine the front end points and rear end points in each of the alternative slices, and using the line connecting the front end point and the rear end point as the anteroposterior diameter of the anterior and posterior midlines;

[0023] An anteroposterior diameter measurement unit for the anterior and posterior midlines to calculate the length and angle of the anteroposterior diameter of the anterior and posterior midlines in each of the alternative slices according to the front end points and rear end points in each of the alternative slices;

[0024] Anterior-posterior midline anteroposterior diameter screening unit, configured to filter the length and angle of the anterior-posterior midline anteroposterior diameter in all the candidate slices to obtain an effective anterior-posterior midline anteroposterior diameter;

[0025] Left and right lung anteroposterior diameter measurement unit, configured to obtain the left lung anteroposterior diameter and the right lung anteroposterior diameter based on the effective anterior-posterior midline anteroposterior diameter and in combination with preset pixel pitch information.

[0026] Optionally, the anterior-posterior midline anteroposterior diameter determination unit includes:

[0027] Region of interest determination subunit, configured to extract an upper preset range rectangular region and a lower preset range rectangular region in the middle of the left lung and the right lung as regions of interest based on the left and right lung segmentation masks, and determine corresponding regions of interest in the bone segmentation mask;

[0028] Anterior-posterior midline anteroposterior diameter determination subunit, configured to extract the midpoint of the lower boundary of the bone segmentation mask in the upper preset range rectangular region as the front endpoint, extract the midpoint of the upper boundary of the bone segmentation mask in the lower preset range rectangular region as the rear endpoint, and use the line connecting the front endpoint and the rear endpoint as the anterior-posterior midline anteroposterior diameter;

[0029] The anterior-posterior midline anteroposterior diameter measurement unit includes:

[0030] Anterior-posterior midline anteroposterior diameter length measurement subunit, configured to calculate the length of the anterior-posterior midline anteroposterior diameter in each candidate slice by using a first preset length calculation formula and based on the front endpoint and the rear endpoint of the anterior-posterior midline anteroposterior diameter in each candidate slice;

[0031] Anterior-posterior midline anteroposterior diameter angle measurement subunit, configured to calculate the angle of the anterior-posterior midline anteroposterior diameter in each candidate slice by using a preset angle calculation formula and based on the front endpoint and the rear endpoint in each candidate slice;

[0032] Wherein, the first preset length calculation formula is:

[0033]

[0034] The preset angle calculation formula is:

[0035]

[0036] In the formula, d ami is the length of the anterior-posterior midline anteroposterior diameter in the i-th candidate slice, θ i is the angle of the anterior-posterior midline anteroposterior diameter in the i-th candidate slice, x tiis the abscissa of the front endpoint of the anteroposterior diameter of the anteroposterior midline in the i-th alternative plane, x bi is the abscissa of the rear endpoint of the anteroposterior diameter of the anteroposterior midline in the i-th alternative plane, y ti is the ordinate of the front endpoint of the anteroposterior diameter of the anteroposterior midline in the i-th alternative plane, y bi is the ordinate of the rear endpoint of the anteroposterior diameter of the anteroposterior midline in the i-th alternative plane.

[0037] Optionally, the anteroposterior midline anteroposterior diameter screening unit includes:

[0038] A mean value calculation sub-unit for calculating the length mean value and the angle mean value of the anteroposterior diameter of the anteroposterior midline in all the alternative planes;

[0039] A mean value elimination sub-unit for eliminating the anteroposterior diameter of the anteroposterior midline whose difference between the length and the length mean value exceeds the length threshold and / or the difference between the angle and the angle mean value exceeds the angle threshold, so as to obtain the effective anteroposterior diameter of the anteroposterior midline in multiple alternative planes;

[0040] The left and right lung anteroposterior diameter measurement unit includes:

[0041] A line segment extraction sub-unit for extracting the longest line segment in the left and right lung segmentation masks that is consistent with the direction of the effective anteroposterior diameter of the anteroposterior midline based on the angles of the effective anteroposterior diameters of the anteroposterior midline in each alternative plane, and respectively serving as the theoretical left lung anteroposterior diameter and the theoretical right lung anteroposterior diameter;

[0042] A left and right lung anteroposterior diameter calculation sub-unit for obtaining the preset coefficient in the preset pixel pitch information, calculating the product of the length of the effective anteroposterior diameter of the anteroposterior midline and the preset coefficient to obtain the length of the physical anteroposterior diameter of the anteroposterior midline, calculating the product of the length of the theoretical left lung anteroposterior diameter and the preset coefficient to obtain the length of the left lung anteroposterior diameter, and calculating the product of the length of the theoretical right lung anteroposterior diameter and the preset coefficient to obtain the length of the right lung anteroposterior diameter.

[0043] Optionally, the transverse diameter positioning module includes:

[0044] A thoracic transverse diameter determination unit for extracting the longest line segment in the left and right lung segmentation masks that is perpendicular to the direction of the effective anteroposterior diameter of the anteroposterior midline based on the angles of the effective anteroposterior diameters of the anteroposterior midline in each alternative plane, serving as the thoracic transverse diameter, and determining the left endpoint and the right endpoint of the thoracic transverse diameter;

[0045] A thoracic transverse diameter length measurement unit for calculating the length of the thoracic transverse diameter in each alternative plane by using a second preset length calculation formula and according to the left endpoint and the right endpoint of the thoracic transverse diameter in each alternative plane;

[0046] A left and right lung transverse diameter calculation unit, configured to extract, based on the angles of the thoracic transverse diameters in each of the alternative slices, the longest line segments in the left and right lung segmentation masks that are consistent with the direction of the thoracic transverse diameter, respectively, as the theoretical left lung transverse diameter and the theoretical right lung transverse diameter, calculate the product of the length of the thoracic transverse diameter and the preset coefficient to obtain the length of the physical thoracic transverse diameter, calculate the product of the length of the theoretical left lung transverse diameter and the preset coefficient to obtain the length of the left lung transverse diameter, and calculate the product of the length of the theoretical right lung transverse diameter and the preset coefficient to obtain the length of the right lung transverse diameter;

[0047] Among them, the second preset length calculation formula is:

[0048]

[0049] In the formula, d ttk is the length of the thoracic transverse diameter in the k-th alternative slice, x lk is the abscissa of the left endpoint of the thoracic transverse diameter in the k-th alternative slice, x rk is the abscissa of the right endpoint of the thoracic transverse diameter in the k-th alternative slice, y lk is the ordinate of the left endpoint of the thoracic transverse diameter in the k-th alternative slice, y rk is the ordinate of the right endpoint of the thoracic transverse diameter in the k-th alternative slice.

[0050] In a second aspect, an embodiment of the present disclosure provides a lung parameter measurement method based on a cross-section of a lung CT image, which is applied to the lung parameter measurement system based on a cross-section of a lung CT image as described in the first aspect. The method includes:

[0051] Obtain a lung CT image through an image acquisition module;

[0052] Process the lung CT image through a left and right lung segmentation module using a first deep learning model or a first image processing algorithm to generate a left and right lung segmentation mask;

[0053] Process the lung CT image through a bone segmentation module using a second deep learning model or a second image processing algorithm to generate a bone segmentation mask;

[0054] Locate the endpoints of the anteroposterior diameter of the anteroposterior midline according to the left and right lung segmentation mask and the bone segmentation mask through an anteroposterior diameter positioning module, and determine the left lung anteroposterior diameter and the right lung anteroposterior diameter according to the anteroposterior diameter of the anteroposterior midline;

[0055] Locate the endpoints of the thoracic transverse diameter according to the left and right lung segmentation mask, the bone segmentation mask and the endpoints of the anteroposterior diameter of the anteroposterior midline through a transverse diameter positioning module, and determine the left lung transverse diameter and the right lung transverse diameter according to the thoracic transverse diameter.

[0056] In a third aspect, an embodiment of the present disclosure provides a computer-readable storage medium storing a computer program, which when executed by a processor, implements the steps of the lung parameter measurement method based on cross-sectional lung CT images described in the second aspect.

[0057] Advantages of the present application:

[0058] The lung parameter measurement system based on cross-sectional lung CT images provided by an embodiment of the present application includes: an image acquisition module for acquiring lung CT images, a left and right lung segmentation module for processing the lung CT images using a first deep learning model or a first image processing algorithm to generate a left and right lung segmentation mask; a bone segmentation module for processing the lung CT images using a second deep learning model or a second image processing algorithm to generate a bone segmentation mask; an anteroposterior diameter positioning module for positioning the endpoints of the anteroposterior diameter of the anteroposterior midline according to the left and right lung segmentation mask and the bone segmentation mask, and determining the anteroposterior diameter of the left lung and the anteroposterior diameter of the right lung according to the anteroposterior diameter of the anteroposterior midline; a transverse diameter positioning module for positioning the endpoints of the thoracic transverse diameter according to the left and right lung segmentation mask, the bone segmentation mask, and the endpoints of the anteroposterior diameter of the anteroposterior midline, and determining the transverse diameter of the left lung and the transverse diameter of the right lung according to the thoracic transverse diameter. The present application improves the reliability of lung development parameter measurement and provides an accurate and consistent parameter measurement method for children's lung development research.

[0059] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following specifically enumerates preferred embodiments and, in conjunction with the accompanying drawings, makes a detailed description as follows. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts. In each drawing, similar components are numbered similarly.

[0061] Figure 1 Shows a schematic structural diagram of a lung parameter measurement device based on cross-sectional lung CT images provided by an embodiment of the present application;

[0062] Figure 2 Shows a schematic diagram of a lung CT image provided by an embodiment of the present application;

[0063] Figure 3 Shows a schematic diagram of a left and right lung segmentation mask provided by an embodiment of the present application;

[0064] Figure 4 Shows a schematic diagram of a bone segmentation mask provided by an embodiment of the present application;

[0065] Figure 5 Shows a schematic diagram of the anteroposterior diameter of the anteroposterior median line and the transverse diameter of the chest provided by an embodiment of the present application;

[0066] Figure 6 Shows a schematic diagram of the theoretical anteroposterior diameter of the left lung and the theoretical anteroposterior diameter of the right lung provided by an embodiment of the present application;

[0067] Figure 7 Shows a schematic diagram of the theoretical transverse diameter of the left lung and the theoretical transverse diameter of the right lung provided by an embodiment of the present application;

[0068] Figure 8 Shows a flowchart of a lung parameter measurement method based on a cross-section of a lung CT provided by an embodiment of the present application.

[0069] Description of main component symbols:

[0070] 100 - Lung parameter measurement system based on cross-section of lung CT; 110 - Image acquisition module; 120 - Left and right lung segmentation module; 121 - Left and right lung segmentation preprocessing unit; 122 - Left and right lung segmentation processing unit; 123 - Left and right lung segmentation postprocessing unit; 130 - Bone segmentation module; 131 - Bone segmentation preprocessing unit; 132 - Bone segmentation processing unit; 133 - Bone segmentation postprocessing unit; 140 - Anteroposterior diameter positioning module; 141 - Alternative layer determination unit; 142 - Anteroposterior diameter determination unit of the anteroposterior median line; 1421 - Region of interest determination subunit; 1422 - Anteroposterior diameter determination subunit of the anteroposterior median line; 143 - Anteroposterior diameter measurement unit of the anteroposterior median line; 1431 - Anteroposterior diameter length measurement subunit; 1432 - Anteroposterior diameter angle measurement subunit; 144 - Anteroposterior diameter screening unit of the anteroposterior median line; 1441 - Mean calculation subunit; 1442 - Mean elimination subunit; 145 - Anteroposterior diameter measurement unit of the left and right lungs; 1451 - Line segment extraction subunit; 1452 - Anteroposterior diameter calculation subunit of the left and right lungs; 150 - Transverse diameter positioning module; 151 - Thoracic transverse diameter determination unit; 152 - Thoracic transverse diameter length measurement unit; 153 - Transverse diameter calculation unit of the left and right lungs. Detailed implementation manners

[0071] The embodiments of the present invention will be described in detail below. The examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions from beginning to end. The embodiments described below by referring to the accompanying drawings are exemplary and are only used to explain the present invention and should not be construed as a limitation to the present invention.

[0072] It should be noted that when an element is referred to as "fixed to" another element, it can be directly on the other element or there may also be an intermediate element. When an element is considered to be "connected" to another element, it can be directly connected to the other element or there may be an intermediate element at the same time. On the contrary, when an element is referred to as being "directly on" another element, there is no intermediate element. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only.

[0073] In the present invention, unless otherwise clearly specified and defined, the terms such as "installed", "connected", "connected", "fixed", etc. should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or integrated; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two elements or the interaction relationship between two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0074] In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present invention, the meaning of "a plurality" is two or more unless otherwise clearly and specifically defined.

[0075] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which this application belongs. The terms used in the description of the template herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0076] Embodiment 1

[0077] As Figure 1 shown, it is a schematic structural diagram of a lung parameter measurement system 100 in an embodiment of the present application. The system includes an image acquisition module 110, a left and right lung segmentation module 120, a bone segmentation module 130, an anteroposterior diameter positioning module 140, and a transverse diameter positioning module 150.

[0078] The image acquisition module 110 is used to acquire lung CT images.

[0079] Understandably, in this embodiment, lung CT (Computed Tomography) images can be acquired through an imaging device. AsFigure 2 As shown, the lung CT image is represented as a grayscale value matrix of size H×W, with the grayscale range being [0, 255].

[0080] The image acquisition module 110 provides the necessary original data for subsequent segmentation and measurement. High-quality CT images can ensure the accuracy and reliability of subsequent processing.

[0081] The left and right lung segmentation module 120 is used to process the lung CT image by using the first deep learning model or the first image processing algorithm to generate a left and right lung segmentation mask.

[0082] Optionally, the left and right lung segmentation module 120 includes a left and right lung segmentation preprocessing unit 121, a left and right lung segmentation processing unit 122, and a left and right lung segmentation postprocessing unit 123.

[0083] The left and right lung segmentation preprocessing unit 121 is used to filter the lung CT image to remove the noise points in the lung CT image and obtain the filtered lung CT image.

[0084] Understandably, the lung CT image is filtered by Gaussian filtering or mean filtering to remove the noise points in the lung CT image and obtain the filtered lung CT image, effectively improving the accuracy of the segmentation model. And the filtered lung CT image is normalized, and the grayscale value range is scaled from [0, 255] to [0, 1], which helps the subsequent deep learning model to converge better.

[0085] The left and right lung segmentation processing unit 122 is used to process the filtered lung CT image by using the first deep learning model or the first image processing algorithm to generate an initial left and right lung segmentation mask.

[0086] Understandably, the filtered lung CT image is processed by using the first deep learning model (such as U-Net or SegFormer, etc.) or the first image processing algorithm (such as threshold method, watershed algorithm) to generate an initial left and right lung segmentation mask M′ of size H×W, and the pixel value is {0, 1, 2}.

[0087] These deep learning models and image processing algorithms can accurately distinguish the background, left lung, and right lung. M′[i, j]=0 represents the background, M′[i, j]=1 represents the left lung area, and M′[i, j]=2 represents the right lung area.

[0088] The left and right lung segmentation postprocessing unit 123 is used to perform morphological operations on the initial left and right lung segmentation mask to remove the noise points and isolated regions in the initial left and right lung segmentation mask and obtain the left and right lung segmentation mask.

[0089] Understandably, morphological operations (such as closing operations) are performed on the initial left and right lung segmentation masks M′ to remove noise points and isolated regions in the initial left and right lung segmentation masks M′, making the segmentation results smoother. Finally, the left and right lung segmentation masks M are obtained, as Figure 3 shown.

[0090] The left and right lung segmentation module 120 realizes the automatic segmentation of the left and right lungs, reduces manual intervention, improves the processing efficiency, and provides accurate left and right lung segmentation masks for subsequent measurements.

[0091] The bone segmentation module 130 is used to process the lung CT image by using a second deep learning model or a second image processing algorithm to generate a bone segmentation mask.

[0092] Optionally, the bone segmentation module 130 includes a bone segmentation preprocessing unit 131, a bone segmentation processing unit 132, and a bone segmentation postprocessing unit 133.

[0093] The bone segmentation preprocessing unit 131 is used to perform histogram equalization on the lung CT image to obtain an enhanced lung CT image.

[0094] Understandably, performing histogram equalization on the lung CT image can also enhance the contrast between bones and soft tissues, which helps to more accurately segment bones and obtain an enhanced lung CT image.

[0095] The bone segmentation processing unit 132 is used to process the enhanced lung CT image by using a second deep learning model or a second image processing algorithm to generate an initial bone segmentation mask.

[0096] Understandably, using a second deep learning model (such as U-Net) or a second image processing algorithm (such as threshold method, watershed algorithm) to process the enhanced lung CT image generates an initial bone segmentation mask S′ with pixel values {0, 1}.

[0097] The deep learning model and the image processing algorithm can accurately distinguish the background and the bone region. S′[i, j]=0 represents the background, and S′[i, j]=1 represents the bone region.

[0098] It should be noted that deep learning models can handle complex image features, while image processing algorithms are applicable to images with higher contrast. In this embodiment, the first deep learning model and the second deep learning model can be the same deep learning model or different deep learning models, and the first image processing algorithm and the second image processing algorithm can be the same image processing algorithm or different image processing algorithms. The deep learning model used can be a two-dimensional segmentation model or a three-dimensional segmentation model, that is, the output can be a two-dimensional numerical matrix or a three-dimensional numerical matrix. A suitable model is selected according to research needs to improve the accuracy of measuring children's lung development parameters. This application embodiment does not make any limitations in this regard.

[0099] The bone segmentation post-processing unit 133 is configured to perform morphological operations on the initial bone segmentation mask to remove noise points and isolated regions in the initial bone segmentation mask S′, and obtain a bone segmentation mask.

[0100] Understandably, morphological operations (such as dilation and erosion) are performed on the initial bone segmentation mask S′ to smooth the bone segmentation edge, remove noise points and isolated regions in the initial bone segmentation mask, and obtain a bone segmentation mask S, as Figure 4 shown.

[0101] The bone segmentation module 130 provides key bone reference information for the subsequent positioning and measurement of the anteroposterior diameter and transverse diameter by accurately segmenting the bone region, thereby improving the measurement accuracy.

[0102] The anteroposterior diameter positioning module 140 is configured to locate the endpoints of the anteroposterior diameter of the anteroposterior midline according to the left and right lung segmentation masks and the bone segmentation mask, and determine the anteroposterior diameter of the left lung and the anteroposterior diameter of the right lung according to the anteroposterior diameter of the anteroposterior midline.

[0103] Optionally, the anteroposterior diameter positioning module 140 includes an alternative slice determination unit 141, an anteroposterior midline anteroposterior diameter determination unit 142, an anteroposterior midline anteroposterior diameter measurement unit 143, an anteroposterior midline anteroposterior diameter screening unit 144, and a left and right lung anteroposterior diameter measurement unit 145.

[0104] The alternative slice determination unit 141 is configured to screen out the slices that contain at least left lung pixels or right lung pixels in the left and right lung segmentation masks as candidate slices, and use the slices within a preset range in the candidate slices as alternative slices.

[0105] Understandably, all slices (slices) of the lung CT image are traversed, and the slices that contain at least left lung pixels or right lung pixels are found in the left and right lung segmentation mask M as candidate slices. Assuming that there are a total of N candidate slices, the slices within the preset range of serial numbers (for example, the middle 20% to 80%) are selected as alternative slices to avoid possible noise interference at the upper and lower edges.

[0106] Anteroposterior diameter determination unit 142 of the anteroposterior midline is configured to determine the front endpoint and the rear endpoint in each alternative plane, and use the line segment connecting the front endpoint and the rear endpoint as the anteroposterior diameter of the anteroposterior midline.

[0107] In an alternative embodiment, the anteroposterior diameter determination unit 142 of the anteroposterior midline includes a region of interest determination subunit 1421 and an anteroposterior diameter determination subunit 1422 of the anteroposterior midline.

[0108] The region of interest determination subunit 1421 is configured to extract the upper preset range rectangular region and the lower preset range rectangular region in the middle of the left lung and the right lung as the regions of interest based on the left and right lung segmentation masks, and determine the corresponding regions of interest in the bone segmentation mask.

[0109] Understandably, in each alternative plane, based on the left and right lung segmentation masks M, as Figure 5 the upper and lower two rectangular frames shown in the figure, extract the upper preset range rectangular region and the lower preset range rectangular region in the middle of the left lung and the right lung (for example, the upper preset range rectangular region refers to 10% higher than the height of the lungs in the image and 60% lower, and the lower preset range rectangular region refers to 10% lower than the height of the lungs in the image and 40% higher) as the regions of interest, and then find the corresponding regions of interest in the bone segmentation mask S.

[0110] The anteroposterior diameter determination subunit 1422 of the anteroposterior midline is configured to extract the midpoint of the lower boundary of the bone segmentation mask in the upper preset range rectangular region as the front endpoint, extract the midpoint of the upper boundary of the bone segmentation mask in the lower preset range rectangular region as the rear endpoint, and use the line segment connecting the front endpoint and the rear endpoint as the anteroposterior diameter of the anteroposterior midline.

[0111] Understandably, in each alternative plane, extract the midpoint of the lower boundary of the bone segmentation mask in the upper preset range rectangular region as the front endpoint, extract the midpoint of the upper boundary of the bone segmentation mask in the lower preset range rectangular region as the rear endpoint, map it to the original image, the front endpoint is represented as (x t , y t ), the rear endpoint is represented as (x b , y b ), and the line segment between the two endpoints is the anteroposterior diameter of the anteroposterior midline, as Figure 5 the dotted line shown in the figure.

[0112] It should be noted that when positioning the anteroposterior diameter of the anteroposterior midline, an AI model can also be used to directly predict the endpoints of the anteroposterior diameter of the anteroposterior midline to improve the efficiency of measuring children's lung development parameters. The embodiments of the present application do not limit this.

[0113] In addition, after determining the anteroposterior diameter of the anterior and posterior midlines, not only can the anteroposterior and transverse diameters of the left and right lungs be measured subsequently, but also the anteroposterior and transverse diameters of finer-grained organ physiological structures such as lung lobes can be measured, increasing the dimensions of the measurement of children's lung development parameters, all of which are within the scope of protection of this application.

[0114] The anteroposterior diameter measurement unit 143 of the anterior and posterior midlines is configured to calculate the length and angle of the anteroposterior diameter of the anterior and posterior midlines in each alternative plane based on the front end point and the rear end point in each alternative plane.

[0115] In an alternative embodiment, the anteroposterior diameter measurement unit 143 of the anterior and posterior midlines includes an anteroposterior diameter length measurement subunit 1431 and an anteroposterior diameter angle measurement subunit 1432.

[0116] The anteroposterior diameter length measurement subunit 1431 is configured to calculate the length of the anteroposterior diameter of the anterior and posterior midlines in each alternative plane by using a first preset length calculation formula and based on the front end point and the rear end point of the anteroposterior diameter of the anterior and posterior midlines in each alternative plane.

[0117] Understandably, by using the first preset length calculation formula and based on the coordinates of the front end point and the rear end point of the anteroposterior diameter of the anterior and posterior midlines in each alternative plane, the length of the anteroposterior diameter of the anterior and posterior midlines in each alternative plane can be calculated. Among them, the first preset length calculation formula is:

[0118]

[0119] In the formula, d ami is the length of the anteroposterior diameter of the anterior and posterior midlines in the i-th alternative plane, x ti is the abscissa of the front end point of the anteroposterior diameter of the anterior and posterior midlines in the i-th alternative plane, x bi is the abscissa of the rear end point of the anteroposterior diameter of the anterior and posterior midlines in the i-th alternative plane, y ti is the ordinate of the front end point of the anteroposterior diameter of the anterior and posterior midlines in the i-th alternative plane, y bi is the ordinate of the rear end point of the anteroposterior diameter of the anterior and posterior midlines in the i-th alternative plane.

[0120] The anteroposterior diameter angle measurement subunit 1432 is configured to calculate the angle of the anteroposterior diameter of the anterior and posterior midlines in each alternative plane by using a preset angle calculation formula and based on the front end point and the rear end point in each alternative plane.

[0121] Understandably, by using the preset angle calculation formula and based on the coordinates of the front end point and the rear end point in each alternative plane, the angle of the anteroposterior diameter of the anterior and posterior midlines in each alternative plane can be calculated. Among them, the preset angle calculation formula is:

[0122]

[0123] where, θ i is the angle of the anteroposterior diameter of the anteroposterior median line in the i-th layer of alternative planes.

[0124] The anteroposterior diameter screening unit 144 of the anteroposterior median line is used to filter the lengths and angles of the anteroposterior diameters of the anteroposterior median line in all alternative planes to obtain an effective anteroposterior diameter of the anteroposterior median line.

[0125] In an alternative embodiment, the anteroposterior diameter screening unit 144 of the anteroposterior median line includes a mean value calculation subunit 1441 and a mean value elimination subunit 1442.

[0126] The mean value calculation subunit 1441 is used to calculate the mean value of the lengths and the mean value of the angles of the anteroposterior diameters of the anteroposterior median line in all alternative planes.

[0127] Understandably, according to the lengths of the anteroposterior diameters of the anteroposterior median line in each alternative plane calculated by the anteroposterior diameter length measurement subunit 1431 and the angles of the anteroposterior diameters of the anteroposterior median line in each alternative plane calculated by the anteroposterior diameter angle measurement subunit 1432, the mean value of the lengths and the mean value of the angles of the anteroposterior diameters of the anteroposterior median line in all alternative planes are calculated.

[0128] The mean value elimination subunit 1442 is used to eliminate the anteroposterior diameters of the anteroposterior median line whose difference between the length and the mean value of the length exceeds the length threshold and / or the difference between the angle and the mean value of the angle exceeds the angle threshold, so as to obtain the effective anteroposterior diameters of the anteroposterior median line in multiple alternative planes.

[0129] Understandably, filtering processing is performed on the lengths and angles of the anteroposterior diameters of the anteroposterior median line in all alternative planes, and the anteroposterior diameters of the anteroposterior median line whose difference from the corresponding mean value exceeds the corresponding threshold are excluded. For example, the anteroposterior diameters of the anteroposterior median line whose difference between the length and the mean value of the length exceeds 20% of the mean value of the length, and / or the anteroposterior diameters of the anteroposterior median line whose difference between the angle and the mean value of the angle exceeds 10 degrees. Finally, the effective values are retained to obtain the effective anteroposterior diameters of the anteroposterior median line in multiple alternative planes.

[0130] The left and right lung anteroposterior diameter measurement unit 145 is used to obtain the left lung anteroposterior diameter and the right lung anteroposterior diameter based on the effective anteroposterior diameter of the anteroposterior median line and in combination with the preset pixel pitch information.

[0131] In an alternative embodiment, the left and right lung anteroposterior diameter measurement unit 145 includes a line segment extraction subunit 1451 and a left and right lung anteroposterior diameter calculation subunit 1452.

[0132] The line segment extraction subunit 1451 is used to extract the longest line segments in the left and right lung segmentation masks that are consistent with the direction of the effective anteroposterior diameter of the anteroposterior median line based on the angles of the effective anteroposterior diameters of the anteroposterior median line in each alternative plane, and use them as the theoretical left lung anteroposterior diameter and the theoretical right lung anteroposterior diameter, such asFigure 6 the dashed line shown in

[0133] Understandably, based on the angular directions of the effective anteroposterior diameter of the anteroposterior midline in each alternative plane, the longest line segments in the left and right lung segmentation masks that are consistent with the direction of the effective anteroposterior diameter of the anteroposterior midline are extracted respectively as the theoretical anteroposterior diameter of the left lung and the theoretical anteroposterior diameter of the right lung, providing key parameters for evaluating the lung morphology.

[0134] The left and right lung anteroposterior diameter calculation subunit 1452 is used to obtain the preset coefficient in the preset pixel pitch information, calculate the product of the length of the effective anteroposterior diameter of the anteroposterior midline and the preset coefficient to obtain the length of the physical anteroposterior diameter of the anteroposterior midline, calculate the product of the length of the theoretical anteroposterior diameter of the left lung and the preset coefficient to obtain the length of the anteroposterior diameter of the left lung, and calculate the product of the length of the theoretical anteroposterior diameter of the right lung and the preset coefficient to obtain the length of the anteroposterior diameter of the right lung.

[0135] Understandably, in combination with the lung CT DICOM (Digital Imaging and Communications in Medicine) data, the DICOM data contains information such as the preset pixel pitch of the image, which is crucial for converting the pixel length into the physical length. Therefore, by extracting the preset coefficient in the preset pixel pitch information and multiplying the preset coefficient by the length of the effective anteroposterior diameter of the anteroposterior midline, the length of the physical anteroposterior diameter of the anteroposterior midline in reality can be obtained. By multiplying the preset coefficient in the preset pixel pitch information by the lengths of the theoretical anteroposterior diameter of the left lung and the theoretical anteroposterior diameter of the right lung respectively, the lengths of the anteroposterior diameter of the left lung and the anteroposterior diameter of the right lung in reality can be further obtained.

[0136] The anteroposterior diameter positioning module 140 effectively excludes outliers through a multi-stage screening strategy, improving the accuracy and reliability of the measurement results. And it can accurately measure the length and angle of the anteroposterior diameter, providing key parameters for evaluating the lung morphology.

[0137] The transverse diameter positioning module 150 is used to locate the endpoints of the thoracic transverse diameter according to the left and right lung segmentation masks, the bone segmentation mask, and the endpoints of the anteroposterior diameter of the anteroposterior midline, and determine the left and right lung transverse diameters according to the thoracic transverse diameter.

[0138] Optionally, the transverse diameter positioning module 150 includes a thoracic transverse diameter determination unit 151, a thoracic transverse diameter length measurement unit 152, and a left and right lung transverse diameter calculation unit 153.

[0139] The thoracic transverse diameter determination unit 151 is used to extract the longest line segment in the left and right lung segmentation masks that is perpendicular to the direction of the effective anteroposterior diameter of the anteroposterior midline based on the angle of the effective anteroposterior diameter of the anteroposterior midline in each alternative plane as the thoracic transverse diameter, and determine the left endpoint and the right endpoint of the thoracic transverse diameter.

[0140] Understandably, among the alternative levels where the anteroposterior diameter of the effective anteroposterior midline is located, based on the angular direction of the anteroposterior diameter of the effective anteroposterior midline, the longest line segment perpendicular to its direction is found in the left and right lung segmentation mask M as the thoracic transverse diameter. The left endpoint of the thoracic transverse diameter is denoted as (x l , y l ), and the right endpoint is denoted as (x r , y r ), as shown by the solid line in Figure 5 .

[0141] The thoracic transverse diameter length measurement unit 152 is configured to calculate the length of the thoracic transverse diameter in each alternative level by using a second preset length calculation formula and based on the left and right endpoints of the thoracic transverse diameter in each alternative level.

[0142] Understandably, by using the second preset length calculation formula and based on the coordinates of the left and right endpoints of the thoracic transverse diameter in each alternative level, the length of the thoracic transverse diameter in each alternative level can be calculated. Among them, the second preset length calculation formula is:[[]]

[0143]

[0144] In the formula, d ttk is the length of the thoracic transverse diameter in the k-th alternative level, x lk is the abscissa of the left endpoint of the thoracic transverse diameter in the k-th alternative level, x rk is the abscissa of the right endpoint of the thoracic transverse diameter in the k-th alternative level, y lk is the ordinate of the left endpoint of the thoracic transverse diameter in the k-th alternative level, and y rk is the ordinate of the right endpoint of the thoracic transverse diameter in the k-th alternative level.

[0145] The left and right lung transverse diameter calculation unit 153 is configured to extract the longest line segments in the left and right lung segmentation masks that are consistent with the direction of the thoracic transverse diameter based on the angle of the thoracic transverse diameter in each alternative level, respectively as the theoretical left lung transverse diameter and the theoretical right lung transverse diameter, and calculate the product of the length of the thoracic transverse diameter and a preset coefficient to obtain the length of the physical thoracic transverse diameter, calculate the product of the length of the theoretical left lung transverse diameter and the preset coefficient to obtain the length of the left lung transverse diameter, and calculate the product of the length of the theoretical right lung transverse diameter and the preset coefficient to obtain the length of the right lung transverse diameter.

[0146] Understandably, based on the angular direction of the thoracic transverse diameter in each alternative level, the longest line segments in the left and right lung segmentation masks that are consistent with the direction of the thoracic transverse diameter are extracted, respectively as the theoretical left lung transverse diameter and the theoretical right lung transverse diameter, providing key parameters for evaluating the lung morphology, as shown by the dashed lines in Figure 7 .

[0147] Next, multiply the preset coefficient in the preset pixel pitch information by the length of the transverse diameter of the chest cavity to obtain the length of the physical transverse diameter of the chest cavity in reality. Multiply the preset coefficient in the preset pixel pitch information by the length of the theoretical left lung transverse diameter and the length of the theoretical right lung transverse diameter respectively to further obtain the length of the left lung transverse diameter and the length of the right lung transverse diameter in reality.

[0148] The transverse diameter positioning module 150 provides key parameters for comprehensively evaluating the lung morphology by measuring the transverse diameter of the chest cavity and the transverse diameters of the left and right lungs. It realizes a fully automated process from the input of lung CT images to the output of the transverse diameter measurement results, improving the processing efficiency.

[0149] The lung parameter measurement system based on the CT cross-section of the lung provided by the embodiment of the present application overcomes the deficiencies in the prior art through modular design, introduction of deep learning, multi-stage screening strategy, direction recognition algorithm, and fully automated processing flow, and has achieved significant improvements in terms of robustness, accuracy, efficiency, etc. The beneficial effects are as follows:

[0150] (1) Higher measurement accuracy and more reliable results: Through precise algorithms and automated processing, the errors caused by manual measurement are reduced, providing accurate parameters for the study of children's lung development;

[0151] (2) Flexible system design and strong scalability: Adopting modular design, each module is relatively independent and can be flexibly combined and expanded according to the needs of children's lung development research;

[0152] (3) Fully automated process and greatly improved efficiency: Realize a fully automated process from the input of lung CT images to the output of parameter measurement results, without manual intervention, improving the efficiency of children's lung development research;

[0153] (4) Still has good performance in complex scenarios: Adopting a multi-stage screening strategy and filters can effectively handle complex situations in children's lung CT images, such as the presence of noise or poor image quality in some parts, ensuring the accuracy of measurement results.

[0154] Embodiment 2

[0155] As Figure 8 shown, it is a flowchart of a lung parameter measurement method based on the CT cross-section of the lung in the embodiment of the present application. The lung parameter measurement method based on the CT cross-section of the lung provided by the embodiment of the present application is applied to the lung parameter measurement system based on the CT cross-section of the lung in Embodiment 1, and specifically includes the following steps:

[0156] Step S110, obtain a lung CT image through the image acquisition module;

[0157] Step S120: Use the first deep learning model or the first image processing algorithm through the left and right lung segmentation module to process the lung CT image, and generate a left and right lung segmentation mask;

[0158] Step S130: Use the second deep learning model or the second image processing algorithm through the bone segmentation module to process the lung CT image, and generate a bone segmentation mask;

[0159] Step S140: Through the anteroposterior diameter positioning module, locate the endpoints of the anteroposterior diameter of the anteroposterior midline according to the left and right lung segmentation masks and the bone segmentation mask, and determine the left lung anteroposterior diameter and the right lung anteroposterior diameter according to the anteroposterior diameter of the anteroposterior midline;

[0160] Step S150: Through the transverse diameter positioning module, locate the endpoints of the thoracic transverse diameter according to the left and right lung segmentation masks, the bone segmentation mask, and the endpoints of the anteroposterior diameter of the anteroposterior midline, and determine the left lung transverse diameter and the right lung transverse diameter according to the thoracic transverse diameter.

[0161] The lung parameter measurement method based on the lung CT cross-section provided by the embodiment of the present application can implement the functions of the lung parameter measurement system corresponding to Embodiment 1, and can achieve the same technical effects. To avoid repetition, it will not be elaborated here.

[0162] The lung parameter measurement method based on the lung CT cross-section provided by the embodiment of the present application improves the reliability of lung development parameter measurement and provides an accurate and consistent parameter measurement method for children's lung development research.

[0163] An embodiment of the present disclosure also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the steps of the lung parameter measurement method based on the lung CT cross-section described in Embodiment 2.

[0164] In several embodiments provided in this application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architectures, functions, and operations of devices, methods, and computer program products according to multiple embodiments of the present invention. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in an alternative implementation, the functions marked in the blocks may occur in a different order from that marked in the accompanying drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, as well as the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system that performs the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0165] As described above, the above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered by the protection scope of the present invention.

Claims

1. A lung parameter measurement system based on lung CT cross-section, characterized in that: The system comprises: An image acquisition module, used for acquiring lung CT images; A left and right lung segmentation module, used to process the lung CT image using a first deep learning model or a first image processing algorithm to generate a left and right lung segmentation mask; A bone segmentation module, used to process the lung CT image using a second deep learning model or a second image processing algorithm to generate a bone segmentation mask; An anteroposterior diameter positioning module, used to locate the endpoints of the anteroposterior diameter of the anteroposterior midline according to the left and right lung segmentation masks and the bone segmentation mask, and determine the anteroposterior diameter of the left lung and the anteroposterior diameter of the right lung according to the anteroposterior diameter of the anteroposterior midline; The transverse diameter positioning module is used to locate the endpoints of the transverse diameter of the thorax according to the left and right lung segmentation masks, the bone segmentation mask and the endpoints of the anterior-posterior diameter of the anterior-posterior midline, and to determine the transverse diameters of the left lung and the right lung according to the transverse diameter of the thorax.

2. The lung parameter measurement system based on lung CT cross-section according to claim 1, characterized in that: The left and right lung segmentation module includes: a left and right lung segmentation preprocessing unit, configured to filter the lung CT image, remove noise points in the lung CT image, obtain a filtered lung CT image, and standardize the filtered lung CT image to obtain a standardized lung CT image; a left and right lung segmentation processing unit, configured to process the standardized lung CT image using the first deep learning model or the first image processing algorithm to generate an initial left and right lung segmentation mask; The left and right lung segmentation post-processing unit is used to perform morphological operations on the initial left and right lung segmentation masks to remove noise points and isolated areas in the initial left and right lung segmentation masks to obtain the left and right lung segmentation masks.

3. The lung parameter measurement system based on lung CT cross-section according to claim 1, characterized in that: The skeleton segmentation module comprises: A bone segmentation preprocessing unit, used for performing histogram equalization on the lung CT image to obtain an enhanced lung CT image; a bone segmentation processing unit, configured to process the enhanced lung CT image using the second deep learning model or the second image processing algorithm to generate an initial bone segmentation mask; The skeleton segmentation post-processing unit is used to perform morphological operations on the initial skeleton segmentation mask to remove noise points and isolated areas in the initial skeleton segmentation mask to obtain the skeleton segmentation mask.

4. The lung parameter measurement system based on lung CT cross-section according to claim 1, characterized in that: The front-rear diameter positioning module comprises: a candidate layer determination unit, configured to select a layer containing at least left lung pixels or right lung pixels in the left and right lung segmentation masks as a candidate layer, and select a layer within a preset range in the candidate layer as a candidate layer; An anteroposterior midline anteroposterior diameter determining unit, used to determine the front end point and the rear end point in each of the candidate layers, and use the line between the front end point and the rear end point as the anteroposterior diameter of the anteroposterior midline; An anteroposterior midline anteroposterior diameter measuring unit, used to calculate the length and angle of the anteroposterior midline anteroposterior diameter in each of the candidate levels according to the front end point and the rear end point in each of the candidate levels; An anteroposterior midline anteroposterior diameter screening unit, used for filtering the length and angle of the anteroposterior midline anteroposterior diameter in all the candidate layers to obtain the effective anteroposterior midline anteroposterior diameter; The left and right lung anterior-posterior diameter measurement unit is used to obtain the left lung anterior-posterior diameter and the right lung anterior-posterior diameter based on the effective anterior-posterior midline anterior-posterior diameter and in combination with preset pixel spacing information.

5. The lung parameter measurement system based on lung CT cross-section according to claim 4, characterized in that: The anterior-posterior midline anterior-posterior diameter determination unit comprises: A region of interest determination subunit is used to extract an upper preset range rectangular area and a lower preset range rectangular area between the left lung and the right lung as a region of interest based on the left and right lung segmentation masks, and determine a corresponding region of interest in the bone segmentation mask; The anterior-posterior midline anteroposterior diameter determination subunit is used to extract the midpoint of the lower boundary of the bone segmentation mask in the upper preset range rectangular area as the front end point, extract the midpoint of the upper boundary of the bone segmentation mask in the lower preset range rectangular area as the rear end point, and use the line between the front end point and the rear end point as the anterior-posterior diameter of the anterior-posterior midline; The anterior-posterior midline anterior-posterior diameter measuring unit comprises: The anteroposterior midline anteroposterior diameter length measuring subunit is used to calculate the length of the anteroposterior diameter of the anteroposterior midline in each of the alternative levels according to the front end point and the rear end point of the anteroposterior midline anteroposterior diameter in each of the alternative levels by using the first preset length calculation formula; The anterior-posterior midline anteroposterior diameter angle measurement subunit is used to calculate the angle of the anterior-posterior diameter of the anterior-posterior midline in each of the alternative levels by using a preset angle calculation formula and according to the front end point and the rear end point in each of the alternative levels; Wherein, the first preset length calculation formula is: The preset angle calculation formula is: Where, d ami is the length of the anterior-posterior diameter of the anterior-posterior midline in the alternative layer of the i-th layer, θ i is the angle of the anteroposterior diameter of the anteroposterior midline in the alternative plane of the i-th layer, x ti is the abscissa of the anterior end point of the anterior-posterior diameter of the anterior-posterior midline in the alternative layer of the i-th layer, x bi is the horizontal coordinate of the posterior end point of the anteroposterior diameter of the anteroposterior midline in the alternative layer of the i-th layer, y ti y is the ordinate of the anterior end point of the anteroposterior diameter of the anteroposterior midline in the alternative layer of the i-th layer, bi It is the ordinate of the posterior end point of the anterior-posterior diameter of the anterior-posterior midline in the alternative layer of the i-th layer.

6. The lung parameter measurement system based on lung CT cross-section according to claim 4, characterized in that: The anterior-posterior midline anterior-posterior diameter screening unit comprises: A mean calculation subunit, used for calculating the length mean and angle mean of the anterior-posterior diameter of the anterior-posterior midline in all the alternative layers; A mean elimination subunit is used to eliminate the anterior-posterior midline anteroposterior diameters whose difference between the length and the length mean exceeds the length threshold and / or whose difference between the angle and the angle mean exceeds the angle threshold, so as to obtain the effective anterior-posterior midline anteroposterior diameters in the plurality of candidate layers; The left and right lung anterior-posterior diameter measurement unit comprises: A line segment extraction subunit is used to extract the longest line segment in the left and right lung segmentation masks that is consistent with the direction of the effective anterior-posterior midline anterior-posterior diameter based on the angle of the effective anterior-posterior midline anterior-posterior diameter in each of the candidate layers, as the theoretical left lung anterior-posterior diameter and the theoretical right lung anterior-posterior diameter respectively; The left and right lung anterior-posterior diameter calculation subunit is used to obtain the preset coefficient in the preset pixel spacing information, calculate the product of the length of the effective anterior-posterior midline anterior-posterior diameter and the preset coefficient, and obtain the length of the physical anterior-posterior midline anterior-posterior diameter, calculate the length of the theoretical left lung anterior-posterior diameter and the product of the preset coefficient, and obtain the length of the left lung anterior-posterior diameter, and calculate the length of the theoretical right lung anterior-posterior diameter and the product of the preset coefficient, and obtain the length of the right lung anterior-posterior diameter.

7. The lung parameter measurement system based on lung CT cross-section according to claim 6, characterized in that: The transverse diameter positioning module comprises: a thorax transverse diameter determination unit, configured to extract, based on the angle of the anteroposterior diameter of the effective anteroposterior midline in each of the candidate layers, the longest line segment perpendicular to the anteroposterior diameter direction of the effective anteroposterior midline in the left and right lung segmentation masks as the thorax transverse diameter, and determine the left endpoint and the right endpoint of the thorax transverse diameter; A thorax transverse diameter length measuring unit, used to calculate the length of the thorax transverse diameter in each of the candidate levels according to the left end point and the right end point of the thorax transverse diameter in each of the candidate levels by using a second preset length calculation formula; A left and right lung transverse diameter calculation unit, for extracting the longest line segment in the left and right lung segmentation masks that is consistent with the direction of the transverse diameter of the thorax based on the angle of the transverse diameter of the thorax in each of the alternative planes, as the theoretical transverse diameter of the left lung and the theoretical transverse diameter of the right lung, respectively, and calculating the product of the length of the transverse diameter of the thorax and the preset coefficient to obtain the length of the physical transverse diameter of the thorax, calculating the product of the length of the theoretical transverse diameter of the left lung and the preset coefficient to obtain the length of the transverse diameter of the left lung, and calculating the product of the length of the theoretical transverse diameter of the right lung and the preset coefficient to obtain the length of the transverse diameter of the right lung; Wherein, the second preset length calculation formula is: Where, d ttk is the length of the transverse diameter of the thorax in the alternative layer of the kth layer, x lk is the abscissa of the left end point of the transverse diameter of the thorax in the alternative layer of the kth layer, x rk is the abscissa of the right end point of the transverse diameter of the thorax in the alternative layer of the kth layer, y lk is the ordinate of the left end point of the transverse diameter of the thorax in the alternative layer of the kth layer, y rk It is the ordinate of the right endpoint of the transverse diameter of the thorax in the alternative layer of the kth layer.

8. A lung parameter measurement method based on lung CT cross-section, characterized in that: Applied to the lung parameter measurement system based on lung CT cross section as described in any one of claims 1 to 7, the method comprising: Acquire lung CT images through an image acquisition module; Processing the lung CT image by using a first deep learning model or a first image processing algorithm through a left and right lung segmentation module to generate a left and right lung segmentation mask; Processing the lung CT image using a second deep learning model or a second image processing algorithm through a bone segmentation module to generate a bone segmentation mask; Locating the endpoints of the anteroposterior diameter of the anterior-posterior midline according to the left and right lung segmentation masks and the bone segmentation mask through the anteroposterior diameter positioning module, and determining the anteroposterior diameter of the left lung and the anteroposterior diameter of the right lung according to the anteroposterior diameter of the anterior-posterior midline; The transverse diameter positioning module locates the endpoints of the transverse diameter of the thorax according to the left and right lung segmentation masks, the bone segmentation mask and the endpoints of the anterior-posterior diameter of the anterior-posterior midline, and determines the transverse diameters of the left lung and the right lung according to the transverse diameter of the thorax.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the lung parameter measurement method based on lung CT cross-section described in claim 8 are implemented.

Images