Lung adhesion detection, analysis method and device, electronic equipment and storage medium

Abstract

The present disclosure relates to a lung adhesion detection, analysis method and device, electronic equipment and storage medium, and relates to the technical field of DR image intelligent diagnosis. The lung adhesion detection method comprises the following steps: respectively extracting the rib edge boundary in the left lung image and / or the rib edge boundary in the right lung image corresponding to a plurality of DR lung images in a deep inhalation state and a deep exhalation state; determining a first motion displacement according to the rib edge boundary in the left lung image corresponding to the plurality of DR lung images; and / or determining a second motion displacement according to the rib edge boundary in the right lung image corresponding to the plurality of DR lung images; and determining whether the left lung and / or the right lung has lung adhesion based on the first motion displacement and / or the second motion displacement and a set motion displacement. The present disclosure can realize lung adhesion detection.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of DR image intelligent diagnosis, and particularly relates to a lung adhesion detection method and device, an electronic device and a storage medium. BACKGROUND

[0002] Digital X-ray (DR) images can provide high-resolution and real-time X-ray images and have been widely applied to bone system, chest, dental and other examinations, such as fracture diagnosis, lung disease screening, dental radiography and the like.

[0003] Pleurisy or chest injury and the like can cause lung adhesion. Specifically, lung adhesion is a phenomenon that the lung and the chest cavity are adhered, especially the lung and the costal margin boundary are adhered, and most of the lung adhesion is caused by pulmonary tuberculosis, chest injury or pleurisy. For patients with pulmonary tuberculosis and chest injury, most of them are likely to have the phenomenon of effusion, at this time, there are also obvious signs, and if the fibrin continues to decline, it is likely to have the phenomenon of adhesion, and even harm, accompanied by granulation tissue production. At the same time, pleural thickening is mainly caused by pleurisy. Due to the failure to find and treat the pleural effusion in time, the pleural effusion stays in the chest cavity for a long time, and the pleural effusion stimulates the pleural effusion and the fibrin attached to the chest cavity wall, so that the chest cavity proliferates and thickens, and further causes lung adhesion.

[0004] At present, as a commonly used and low-cost image device for diagnosing lung diseases, it is necessary to automatically detect lung adhesion based on DR images to assist in improving the diagnosis level of doctors. SUMMARY

[0005] The present disclosure provides a lung adhesion detection and analysis method and device, an electronic device and a storage medium.

[0006] According to an aspect of the present disclosure, a lung adhesion detection method is provided, comprising:

[0007] extracting the costal margin boundary in the left lung image and / or the costal margin boundary in the right lung image corresponding to a plurality of DR lung images in a deep inhalation state and a deep exhalation state, respectively;

[0008] determining a first motion displacement according to the costal margin boundary in the left lung image corresponding to the plurality of DR lung images, and / or determining a second motion displacement according to the costal margin boundary in the right lung image corresponding to the plurality of DR lung images;

[0009] determining whether the left lung and / or the right lung has lung adhesion based on the first motion displacement and / or the second motion displacement and a set motion displacement, respectively.

[0010] Preferably, the method for extracting the rib edge boundary in the left lung image corresponding to the plurality of DR lung images in the deep inspiration state and the deep expiration state respectively comprises:

[0011] A direction derivative template of the left chest image corresponding to the plurality of DR lung images is constructed by using the direction derivative, and a set weighted depth of the direction derivative template is set;

[0012] The left chest image is subjected to template traversal of the direction derivative by using the direction derivative template corresponding to the set weighted depth, and a result of the template traversal is superimposed into the left chest image to obtain a left chest superimposed image;

[0013] The left chest superimposed image is subjected to binarization processing to obtain a left rib edge binary image;

[0014] A left rib edge angle image to be screened is obtained according to the left rib edge binary image and the left chest superimposed image;

[0015] A screened left rib edge angle image is obtained based on the left rib edge angle image to be screened and a first set rib edge angle;

[0016] The left rib edge angle image is subjected to connected domain selection to obtain a rib edge boundary in the left lung image corresponding to a maximum connected domain; and / or,

[0017] The method for extracting the second motion displacement of the rib edge boundary in the right lung image corresponding to the plurality of DR lung images in the deep inspiration state and the deep expiration state respectively comprises:

[0018] A direction derivative template of the right chest image corresponding to the plurality of DR lung images is constructed by using the direction derivative, and a set weighted depth of the direction derivative template is set;

[0019] The right chest image is subjected to template traversal of the direction derivative by using the direction derivative template corresponding to the set weighted depth, and a result of the template traversal is superimposed into the right chest image to obtain a right chest superimposed image;

[0020] The right chest superimposed image is subjected to binarization processing to obtain a right rib edge binary image;

[0021] A right rib edge angle image to be screened is obtained according to the right rib edge binary image and the right chest superimposed image;

[0022] A screened right rib edge angle image is obtained based on the right rib edge angle image to be screened and a second set rib edge angle;

[0023] The right rib edge angle image is subjected to connected domain selection to obtain a rib edge boundary in the right lung image corresponding to a maximum connected domain.

[0024] Preferably, the method for obtaining the rib angle image of the left side to be screened according to the left rib edge binary image and the left chest superimposed image comprises:

[0025] The morphological opening-closing operation and thinning processing are performed on the left rib edge binary image to obtain a morphologically processed left rib edge binary image.

[0026] The AND operation is performed on the gradient direction angle of each pixel in the morphologically processed left rib edge binary image and the left chest superimposed image to obtain the rib angle image of the left side to be screened; and / or,

[0027] The method for obtaining the rib angle image of the right side to be screened according to the right rib edge binary image and the right chest superimposed image comprises:

[0028] The morphological opening-closing operation and thinning processing are performed on the right rib edge binary image to obtain a morphologically processed right rib edge binary image.

[0029] The AND operation is performed on the gradient direction angle of each pixel in the morphologically processed right rib edge binary image and the right chest superimposed image to obtain the rib angle image of the right side to be screened; and / or,

[0030] Before the direction derivative template of the left chest image corresponding to the plurality of DR lung images is constructed by using the direction derivative, the left chest image is subjected to a set scale Gaussian blur to obtain a corresponding left chest Gaussian blur image.

[0031] Further, the direction derivative template of the left chest Gaussian blur image is constructed by using the direction derivative.

[0032] In the rib boundary process of the left chest image, the left chest Gaussian blur image is subjected to template traversal of the direction derivative by using the direction derivative template corresponding to the set weighting depth, and the result of the template traversal is superimposed into the left chest image to obtain a left chest Gaussian blur superimposed image.

[0033] The left chest Gaussian blur superimposed image is subjected to a binaryzation processing to obtain a left rib edge binary image; the rib angle image of the left side to be screened is obtained according to the left rib edge binary image and the left chest Gaussian blur superimposed image; and / or,

[0034] Before the direction derivative template of the right chest image corresponding to the plurality of DR lung images is constructed by using the direction derivative, the right chest image is subjected to a set scale Gaussian blur to obtain a corresponding right chest Gaussian blur image.

[0035] Further, the direction derivative template of the right chest Gaussian blur image is constructed by using the direction derivative.

[0036] In the rib edge boundary process of the right chest image, a directional derivative template corresponding to the set weighted depth is used to perform template traversal of the right chest Gaussian blur image, and a result of the template traversal is superimposed into the right chest image to obtain a right chest Gaussian blur superimposed image.

[0037] The right chest Gaussian blur superimposed image is subjected to a binarization process to obtain a right rib edge binary image; and a right rib edge angle image to be screened is obtained according to the right rib edge binary image and the right chest Gaussian blur superimposed image.

[0038] Preferably, the method for respectively extracting rib edge boundaries in left lung images and / or rib edge boundaries in right lung images corresponding to the plurality of DR lung images in the deep inspiration state and the deep expiration state comprises:

[0039] The plurality of DR lung images in the deep inspiration state and the deep expiration state are respectively subjected to left lung and right lung segmentation to obtain left lung images and / or right lung images corresponding to the plurality of DR lung images;

[0040] The left lung images and / or the right lung images are respectively subjected to rib edge boundary detection to obtain rib edge boundaries in left lung images and / or rib edge boundaries in right lung images corresponding to the plurality of DR lung images; and / or,

[0041] The method for respectively extracting rib edge boundaries in left lung images and / or rib edge boundaries in right lung images corresponding to the plurality of DR lung images comprises:

[0042] A plurality of left lung contour lines and a plurality of right lung contour lines corresponding to the left lung images and / or the right lung images corresponding to the plurality of DR lung images are extracted;

[0043] The plurality of left lung contour lines and / or the plurality of right lung contour lines are respectively used to determine rib edge boundaries in left lung images and / or rib edge boundaries in right lung images corresponding to the plurality of DR lung images; and / or,

[0044] The method for respectively extracting rib edge boundaries in left lung images and / or rib edge boundaries in right lung images corresponding to the plurality of DR lung images comprises:

[0045] The plurality of left lung contour lines and / or the plurality of right lung contour lines are respectively used to determine a plurality of left lung motion peaks, a plurality of left lung motion troughs, and / or a plurality of right lung motion peaks, and a plurality of right lung motion troughs corresponding thereto;

[0046] determine the rib boundary in the left lung image corresponding to the plurality of DR lung images based on the plurality of left lung movement peaks and the plurality of left lung movement troughs; and / or, determine the rib boundary in the right lung image corresponding to the plurality of DR lung images based on the plurality of right lung movement peaks and the plurality of right lung movement troughs; and / or,

[0047] The method for determining the rib boundary in the left lung image corresponding to the plurality of DR lung images based on the plurality of left lung movement peaks and the plurality of left lung movement troughs comprises:

[0048] connecting the plurality of left lung movement peaks and the plurality of left lung movement troughs corresponding thereto to obtain a plurality of first division lines;

[0049] configuring the rib boundary in the left lung image on the right side of the plurality of first division lines as the rib boundary in the left lung image corresponding to the plurality of DR lung images; and / or,

[0050] The method for determining the rib boundary in the right lung image corresponding to the plurality of DR lung images based on the plurality of right lung movement peaks and the plurality of right lung movement troughs comprises:

[0051] connecting the plurality of right lung movement peaks and the plurality of right lung movement troughs corresponding thereto to obtain a plurality of second division lines;

[0052] configuring the rib boundary in the right lung image on the left side of the plurality of second division lines as the rib boundary in the right lung image corresponding to the plurality of DR lung images; and / or,

[0053] The method for performing left lung and right lung segmentation on the plurality of DR lung images in the deep inhalation state and the deep exhalation state respectively to obtain the left lung image and the right lung image corresponding to the plurality of DR lung images comprises:

[0054] performing rib boundary, lung apex boundary, and mediastinum and diaphragm edge detection on the left chest image and the right chest image of the plurality of DR lung images in the deep inhalation state and the deep exhalation state respectively to obtain the left lung image and the right lung image corresponding to the plurality of DR lung images; or,

[0055] obtaining a segmentation model of a preset convolutional neural network and a DR lung region label image used for training the segmentation model; and training the segmentation model using the DR lung region label image used for training the segmentation model;

[0056] performing left lung and right lung segmentation on the plurality of DR lung images in the deep inhalation state and the deep exhalation state based on the trained segmentation model to obtain the left lung image and the right lung image corresponding to the plurality of DR lung images.

[0057] Preferably, the method for determining the first motion displacement according to the rib edge boundary in the left lung image corresponding to the plurality of DR lung images comprises:

[0058] The rib edge boundaries in the left lung images corresponding to the plurality of DR lung images are registered to obtain left lung registered rib edge boundaries;

[0059] The first motion displacement is determined according to the left lung registered rib edge boundaries; and / or,

[0060] The method for determining the first motion displacement according to the left lung registered rib edge boundaries comprises: calculating a plurality of first distances corresponding to each boundary point in the left lung registered rib edge boundaries and the left lung rib edge boundaries before and after registration based on the left lung registered rib edge boundaries; and configuring the plurality of first distances as the first motion displacement; or, the method for determining the first motion displacement according to the left lung registered rib edge boundaries comprises: dividing the left lung registered rib edge boundaries and the left lung rib edge boundaries before registration respectively based on a plurality of positions of the left lung ribs or a plurality of set positions to obtain a first plurality of segmented left lung rib edge boundaries before registration and a first plurality of segmented left lung rib edge boundaries after registration; calculating a plurality of first distances corresponding to each boundary point in the first plurality of segmented left lung rib edge boundaries before registration and the first plurality of segmented left lung rib edge boundaries after registration based on the left lung registered rib edge boundaries, and configuring the plurality of first distances as the first motion displacement; and / or,

[0061] The method for determining the plurality of positions of the left lung ribs comprises: performing rib detection on the left lung registered rib edge boundaries and the left lung rib edge boundaries before registration respectively to obtain a plurality of positions of the left lung ribs; and / or,

[0062] The method for performing rib detection on the left lung registered rib edge boundaries and the left lung rib edge boundaries before registration respectively comprises:

[0063] A left lung rib segmentation model corresponding to a preset convolutional neural network is obtained;

[0064] The left lung rib segmentation model is used to perform rib detection on the lung registered rib edge boundaries and the left lung rib edge boundaries before registration to obtain a plurality of positions of the left lung ribs; and / or,

[0065] Before the left lung rib segmentation model is used to perform rib detection on the lung registered rib edge boundaries and the left lung rib edge boundaries before registration to obtain a plurality of positions of the left lung ribs, a left lung rib label corresponding to the left lung ribs is obtained, and the left lung rib segmentation model is trained based on the left lung rib label; the left lung rib segmentation model after training is used to perform rib detection on the lung registered rib edge boundaries and the left lung rib edge boundaries before registration to obtain a plurality of positions of the left lung ribs; and / or,

[0066] The method for determining the plurality of positions of the left lung ribs comprises: performing rib detection on the left lung registered costophrenic boundary and the left lung costophrenic boundary before registration respectively, to obtain the plurality of positions of the left lung ribs; and / or,

[0067] The method for performing rib detection on the left lung registered costophrenic boundary and the left lung costophrenic boundary before registration respectively comprises:

[0068] Obtaining a left lung rib segmentation model corresponding to a preset convolutional neural network;

[0069] Performing rib detection on the lung registered costophrenic boundary and the left lung costophrenic boundary before registration based on the left lung rib segmentation model, to obtain the plurality of positions of the left lung ribs

[0070] The method for determining the second motion displacement according to the costophrenic boundary in the right lung image corresponding to the plurality of DR lung images comprises:

[0071] Performing registration on the costophrenic boundary in the right lung image corresponding to the plurality of DR lung images, to obtain a right lung registered costophrenic boundary;

[0072] Determining the second motion displacement according to the right lung registered costophrenic boundary; and / or,

[0073] The method for determining the second motion displacement according to the right lung registered costophrenic boundary comprises:

[0074] Based on the right lung registered costophrenic boundary, calculating a plurality of second distances corresponding to each boundary point in the right lung costophrenic boundary before registration and after registration, and configuring the plurality of second distances as the second motion displacement; or, the method for determining the second motion displacement according to the right lung registered costophrenic boundary comprises: based on the plurality of positions of the right lung ribs or setting a plurality of positions, dividing the right lung costophrenic boundary before registration and after registration respectively, to obtain a second plurality of segments of the right lung costophrenic boundary before registration and a second plurality of segments of the right lung costophrenic boundary after registration; based on the right lung registered costophrenic boundary, calculating a plurality of second distances corresponding to each boundary point in the second plurality of segments of the right lung costophrenic boundary before registration and the second plurality of segments of the right lung costophrenic boundary after registration respectively, and configuring the plurality of second distances as the second motion displacement; and / or,

[0075] The method for determining the plurality of positions of the right lung ribs comprises: performing rib detection on the right lung registered costophrenic boundary and the right lung costophrenic boundary before registration respectively, to obtain the plurality of positions of the right lung ribs; and / or,

[0076] The method for performing rib detection on the right lung registered costophrenic boundary and the right lung costophrenic boundary before registration respectively comprises:

[0077] Obtaining a right lung rib segmentation model corresponding to a preset convolutional neural network;

[0078] based on the right lung rib segmentation model, rib detection is performed on the lung registered costophrenic border and the pre-registration right lung costophrenic border to obtain a plurality of positions of the right lung ribs; and / or,

[0079] Before the rib detection based on the right lung rib segmentation model, a right lung rib label corresponding to the right lung ribs is obtained, and the right lung rib segmentation model is trained based on the right lung rib label; based on the trained right lung rib segmentation model, rib detection is performed on the lung registered costophrenic border and the pre-registration right lung costophrenic border to obtain a plurality of positions of the right lung ribs; and / or,

[0080] The method for determining whether the left lung has lung adhesion based on the first motion displacement and the set motion displacement respectively, comprises:

[0081] If the first motion displacement is less than or equal to the set motion displacement, it is determined that the left lung has lung adhesion; otherwise, the left lung does not have lung adhesion; and / or,

[0082] The method for determining whether the right lung has lung adhesion based on the second motion displacement and the set motion displacement respectively, comprises:

[0083] If the second motion displacement is less than or equal to the set motion displacement, it is determined that the right lung has lung adhesion; otherwise, the right lung does not have lung adhesion.

[0084] According to an aspect of the present disclosure, an analysis method is provided, which comprises or applies the lung adhesion detection method as described above to determine that the left lung has lung adhesion and / or the right lung has lung adhesion; and,

[0085] It is determined whether there is a functional disorder in the diaphragm movement of the patient corresponding to the plurality of DR lung images in the deep inspiration state and the deep expiration state;

[0086] If there is no functional disorder, a final analysis result corresponding to that the left lung has lung adhesion and / or the right lung has lung adhesion is given; otherwise, a final analysis result corresponding to that the left lung does not have lung adhesion and / or the right lung does not have lung adhesion is given.

[0087] According to an aspect of the present disclosure, a lung adhesion detection device is provided, which comprises: an extraction unit configured to extract a costophrenic border in a left lung image and / or a costophrenic border in a right lung image corresponding to a plurality of DR lung images in a deep inspiration state and a deep expiration state respectively;

[0088] a displacement determination unit configured to determine a first motion displacement based on a rib edge boundary in a left lung image corresponding to the plurality of DR lung images, and / or determine a second motion displacement based on a rib edge boundary in a right lung image corresponding to the plurality of DR lung images;

[0089] a detection unit configured to determine whether the left lung and / or the right lung has lung adhesion based on the first motion displacement and / or the second motion displacement and a set motion displacement, respectively.

[0090] According to an aspect of the present disclosure, there is provided an analysis device comprising: or applying the lung adhesion detection method as described above or the lung adhesion detection device as described above to determine whether the left lung has lung adhesion and / or the right lung has lung adhesion; and,

[0091] an evaluation unit configured to evaluate whether there is a functional disorder in diaphragm movement of the patient corresponding to the plurality of DR lung images in the deep inhalation state and the deep exhalation state;

[0092] an analysis unit configured to give a final analysis result corresponding to whether the left lung has lung adhesion and / or the right lung has lung adhesion if there is no functional disorder, or give a final analysis result corresponding to whether the left lung has no lung adhesion and / or the right lung has no lung adhesion if there is a functional disorder.

[0093] According to an aspect of the present disclosure, there is provided an electronic device comprising:

[0094] a processor;

[0095] a memory for storing processor-executable instructions;

[0096] wherein the processor is configured to execute the lung adhesion detection method as described above and / or the analysis method as described above.

[0097] According to an aspect of the present disclosure, there is provided a computer-readable storage medium having stored thereon computer program instructions, which, when executed by a processor, implement the lung adhesion detection method as described above and / or the analysis method as described above.

[0098] In the embodiments of the present disclosure, a lung adhesion detection and analysis method and device, an electronic device, and a storage medium are provided, which can realize automatic detection of lung adhesion based on DR images, thereby assisting in improving the diagnosis level of physicians, and further solving the problem that the intelligent auxiliary diagnosis level of DR images is low and lung adhesion cannot be automatically detected.

[0099] It should be understood that the foregoing general description and the following detailed description are only exemplary and explanatory, rather than limiting the present disclosure.

[0100] Other features and aspects of the present disclosure will become apparent from the following detailed description, taken in conjunction with the accompanying drawings, which illustrate, by way of example, the principles of the disclosure. Attached Figure Description

[0101] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the specification, serve to illustrate the technical solutions of this disclosure.

[0102] Figure 1 A flowchart of a lung adhesion detection method according to an embodiment of the present disclosure is shown;

[0103] Figure 2 This is a block diagram illustrating an electronic device 800 according to an exemplary embodiment;

[0104] Figure 3 This is a block diagram illustrating an electronic device 1900 according to an exemplary embodiment. Detailed Implementation

[0105] Various exemplary embodiments, features, and aspects of this disclosure will now be described in detail with reference to the accompanying drawings. The same reference numerals in the drawings denote elements that have the same or similar functions. Although various aspects of the embodiments are shown in the drawings, they are not necessarily drawn to scale unless specifically indicated otherwise.

[0106] The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments.

[0107] In this document, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent three cases: A alone, A and B simultaneously, and B alone. Furthermore, the term "at least one" in this document means any combination of at least two of any one or more elements. For example, including at least one of A, B, and C can mean including any one or more elements selected from the set consisting of A, B, and C.

[0108] Furthermore, to better illustrate this disclosure, numerous specific details are set forth in the following detailed description. Those skilled in the art will understand that this disclosure can be practiced without certain specific details. In some instances, methods, means, components, and circuits well known to those skilled in the art have not been described in detail in order to highlight the main points of this disclosure.

[0109] It is understood that the various embodiments of lung adhesion detection methods based on DR images mentioned in this disclosure can be combined with each other to form combined embodiments without violating the principle and logic. Due to space limitations, this disclosure will not elaborate further.

[0110] Further, the disclosure also provides a lung adhesion detection device based on a DR image, an electronic device, a computer readable storage medium, and a program, which can be used to implement any of the lung adhesion detection methods based on a DR image provided by the disclosure. The corresponding technical solutions and descriptions are described in the method part and are not repeated here.

[0111] Figure 1 A flowchart of the lung adhesion detection method based on a DR image according to an embodiment of the disclosure is shown as follows. Figure 1 As shown, the lung adhesion detection method based on a DR image includes the following steps. Step S101: Extracting rib edge boundaries in left lung images and / or rib edge boundaries in right lung images corresponding to a plurality of DR lung images in a deep inhalation state and a deep exhalation state, respectively. Step S102: Determining a first motion displacement according to the rib edge boundaries in the left lung images corresponding to the plurality of DR lung images, and / or determining a second motion displacement according to the rib edge boundaries in the right lung images corresponding to the plurality of DR lung images. Step S103: Determining whether lung adhesion exists in the left lung and / or the right lung based on the first motion displacement and / or the second motion displacement and a set motion displacement, respectively. This can assist in improving the diagnosis level of a physician, thereby solving the problem that the intelligent auxiliary diagnosis level of a DR image is low and cannot automatically detect lung adhesion.

[0112] Step S101: Extracting rib edge boundaries in left lung images and / or rib edge boundaries in right lung images corresponding to a plurality of DR lung images in a deep inhalation state and a deep exhalation state, respectively.

[0113] In the embodiments and other possible embodiments of the disclosure, a digital X-ray (DR) image can provide a high-resolution and real-time X-ray image and has been widely applied to examinations of skeletal systems, chests, dentistry, etc., such as fracture diagnosis, lung disease screening, and dental radiography. Therefore, a DR image device can be used to image skeletal systems, chests, dentistry, etc.

[0114] In the embodiments and other possible embodiments of the disclosure, the plurality of DR lung images in the deep inhalation state and the deep exhalation state are configured as one DR lung image taken in a deep inhalation state and one DR lung image taken in a deep exhalation state. The one DR lung image taken in the deep inhalation state corresponds to rib edge boundaries in a left lung image and / or rib edge boundaries in a right lung image, and the one DR lung image taken in the deep exhalation state corresponds to rib edge boundaries in a left lung image and / or rib edge boundaries in a right lung image.

[0115] In the embodiments of the present disclosure, the method for respectively performing rib edge boundary, lung apex boundary and mediastinum and diaphragm edge detection on a left chest image and a right chest image of each of the plurality of DR lung region images to obtain a DR lung region label image corresponding to the plurality of DR lung region images comprises: obtaining a left lung segmentation image corresponding to the DR lung region label image according to the rib edge boundary, the lung apex boundary and the mediastinum and diaphragm edge of the left chest image; and obtaining a right lung segmentation image corresponding to the DR lung region label image according to the rib edge boundary, the lung apex boundary and the mediastinum and diaphragm edge of the right chest image.

[0116] In the embodiments of the present disclosure, the method for performing rib edge boundary on the left chest image comprises: constructing a directional derivative template of the left chest image by using a directional derivative, and setting a set weighting depth of the directional derivative template; performing template traversal of the directional derivative on the left chest image by using the directional derivative template corresponding to the set weighting depth, and superimposing a result of the template traversal into the left chest image to obtain a left chest superimposed image; performing binaryzation processing on the left chest superimposed image to obtain a left rib edge binary image; obtaining a left rib edge angle image to be screened according to the left rib edge binary image and the left chest superimposed image; obtaining a screened left rib edge angle image based on the left rib edge angle image to be screened and a first set rib edge angle; and performing connected domain selection on the left rib edge angle image to obtain a rib edge boundary corresponding to a maximum connected domain.

[0117] In the embodiments of the present disclosure, the method for obtaining the left rib edge angle image to be screened according to the left rib edge binary image and the left chest superimposed image comprises: performing morphological opening and closing operation and thinning processing on the left rib edge binary image to obtain a morphologically processed left rib edge binary image; and performing AND operation on a gradient direction angle of each pixel in the morphologically processed left rib edge binary image and the left chest superimposed image to obtain the left rib edge angle image to be screened.

[0118] In the embodiments of the present disclosure, before the directional derivative template of the left chest image is constructed by using the directional derivative, a set scale Gaussian blur is performed on the left chest image to obtain a corresponding left chest Gaussian blur image; then, the directional derivative template of the left chest Gaussian blur image is constructed by using the directional derivative; in the process of performing rib edge boundary on the left chest image, the directional derivative template traversal is performed on the left chest Gaussian blur image by using the directional derivative template corresponding to the set weighting depth, and a result of the template traversal is superimposed into the left chest image to obtain a left chest Gaussian blur superimposed image; the binaryzation processing is performed on the left chest Gaussian blur superimposed image to obtain a left rib edge binary image; and the left rib edge angle image to be screened is obtained according to the left rib edge binary image and the left chest Gaussian blur superimposed image.

[0119] In the embodiment of the present disclosure, the method for performing costal margin boundary on the right chest image comprises: constructing a direction derivative template of the right chest image by using a direction derivative, and setting a set weighting depth of the direction derivative template; performing template traversal of the direction derivative on the right chest image by using the direction derivative template corresponding to the set weighting depth, superimposing a result of the template traversal into the right chest image to obtain a right chest superimposed image; performing binaryzation processing on the right chest superimposed image to obtain a right costal margin binary image; obtaining a right costal margin angle image to be screened according to the right costal margin binary image and the right chest superimposed image; obtaining a screened right costal margin angle image based on the right costal margin angle image to be screened and a second set costal margin angle; and performing connected domain selection on the right costal margin angle image to obtain a costal margin boundary corresponding to a maximum connected domain.

[0120] In the embodiment of the present disclosure, the method for obtaining the right costal margin angle image to be screened according to the right costal margin binary image and the right chest superimposed image comprises: performing morphological opening-closing operation and thinning processing on the right costal margin binary image to obtain a morphologically processed right costal margin binary image; and performing AND operation on a gradient direction angle of each pixel in the morphologically processed right costal margin binary image and the right chest superimposed image to obtain the right costal margin angle image to be screened.

[0121] In the embodiment of the present disclosure, before the direction derivative template of the right chest image is constructed by using the direction derivative, a set scale Gaussian blur is performed on the right chest image to obtain a corresponding right chest Gaussian blur image; then, the direction derivative template of the right chest Gaussian blur image is constructed by using the direction derivative; in the process of performing costal margin boundary on the right chest image, the direction derivative template corresponding to the set weighting depth is used to perform template traversal of the direction derivative on the right chest Gaussian blur image, and a result of the template traversal is superimposed into the right chest image to obtain a right chest Gaussian blur superimposed image; binaryzation processing is performed on the right chest Gaussian blur superimposed image to obtain a right costal margin binary image; and a right costal margin angle image to be screened is obtained according to the right costal margin binary image and the right chest Gaussian blur superimposed image.

[0122] In the embodiment of the present disclosure, the method for performing left lung apex boundary detection on the left chest image comprises: determining a left lung apex detection region according to the left chest image; determining a left lung apex edge binary image according to the left lung apex detection region; and obtaining a fitted left lung apex boundary by using quadratic function fitting according to the left lung apex edge binary image.

[0123] In embodiments of the present disclosure, the method for determining a left lung apex detection region from the left chest image comprises: detecting a first coordinate corresponding to the uppermost coordinate point of the rib margin of the left chest image; and configuring a region formed by the first coordinate and the uppermost coordinate point of the rightmost corner of the left chest image as the left lung apex detection region.

[0124] In embodiments of the present disclosure, the method for detecting a right lung apex boundary from the right chest image comprises: determining a right lung apex detection region from the right chest image; determining a right lung apex edge binary image from the right lung apex detection region; and obtaining a fitted right lung apex boundary by using a quadratic function fitting based on the right lung apex edge binary image.

[0125] In embodiments of the present disclosure, the method for determining a right lung apex detection region from the right chest image comprises: detecting a second coordinate corresponding to the uppermost coordinate point of the rib margin of the right chest image; and configuring a region formed by the second coordinate and the uppermost coordinate point of the leftmost corner of the left chest image as the right lung apex detection region.

[0126] In embodiments of the present disclosure, the method for detecting a left lung mediastinum and transverse septum edge from the left chest image comprises: performing binaryzation processing on the left chest image to obtain a left chest binary image; performing edge detection on the left chest binary image to obtain a left chest edge binary image; obtaining a left chest edge angle image based on the gradient direction angle of each pixel in the left chest binary image and the left chest edge binary image; obtaining a selected left transverse septum and mediastinum edge angle image based on the left chest edge angle image and the edge angle range of the transverse septum and mediastinum; and obtaining a left lung mediastinum and transverse septum edge corresponding to the maximum connected domain by performing connected domain selection processing based on the selected left transverse septum and mediastinum edge angle image.

[0127] In embodiments of the present disclosure, the method for performing binaryzation processing on the left chest image to obtain a left chest binary image comprises: performing contrast enhancement processing and maximum inter-class variance processing on a left chest Gaussian blur image corresponding to the left chest image to obtain a left chest binary image.

[0128] In embodiments of the present disclosure, the method for determining the gradient direction angle of each pixel in the left chest edge binary image comprises: performing horizontal gradient and vertical gradient calculation on a left chest Gaussian blur image corresponding to the left chest image to obtain a horizontal gradient image and a vertical gradient image of the left chest; and obtaining the gradient direction angle of each pixel in the left chest Gaussian blur image based on the horizontal gradient image and the vertical gradient image of the left chest.

[0129] In the embodiments of the present disclosure, the method for right lung mediastinum and transverse septum edge detection on the right chest image respectively comprises: performing binaryzation processing on the right chest image to obtain a right chest binary image; performing edge detection on the right chest binary image to obtain a right chest edge binary image; obtaining a right chest edge angle image according to the gradient direction angle of each pixel in the right chest binary image and the right chest edge binary image; obtaining a selected right transverse septum and mediastinum edge angle image according to the obtained right chest edge angle image and the edge angle range of the set transverse septum and mediastinum; and obtaining the right lung mediastinum and transverse septum edge corresponding to the maximum connected domain through connected domain selection processing according to the selected right transverse septum and mediastinum edge angle image.

[0130] In the embodiments of the present disclosure, the method for performing binaryzation processing on the right chest image to obtain a right chest binary image comprises: performing contrast enhancement processing and maximum inter-class variance processing on a right chest Gaussian blur image corresponding to the right chest image to obtain a right chest binary image.

[0131] In the embodiments of the present disclosure, the method for determining the gradient direction angle of each pixel in the right chest edge binary image comprises: performing horizontal gradient and vertical gradient calculation on a right chest Gaussian blur image corresponding to the right chest image to obtain a horizontal gradient image and a vertical gradient image of the right chest; and obtaining the gradient direction angle of each pixel in the right chest Gaussian blur image based on the horizontal gradient image and the vertical gradient image of the right chest.

[0132] In the embodiments of the present disclosure, the method for obtaining a left lung segmentation image according to the rib edge boundary, the lung apex boundary, and the mediastinum and transverse septum edge of the left chest image comprises: calculating a first left lung apex edge point and a left lung mediastinum edge point corresponding to the shortest distance between the lung apex boundary and the mediastinum edge in the left chest image; calculating a second left lung apex edge point and a first left lung rib edge point corresponding to the shortest distance between the lung apex boundary and the rib edge in the left chest image; calculating a second rib edge point and a diaphragm edge point corresponding to the shortest distance between the rib edge boundary and the transverse septum edge in the left chest image; and obtaining a left lung segmentation image based on the first left lung apex edge point, the left lung mediastinum edge point, the second left lung apex edge point, the first left lung rib edge point, the second rib edge point, and the diaphragm edge point.

[0133] In the embodiments of the present disclosure, the method for obtaining the right lung segmentation image according to the costophrenic boundary, the apical boundary and the longitudinal and transverse mediastinal edges corresponding to the right chest image comprises: calculating a first right lung apical edge point and a right lung longitudinal mediastinal edge point corresponding to the shortest distance between the apical boundary and the longitudinal mediastinal edge in the right chest image; calculating a second right lung apical edge point and a first right lung costophrenic edge point corresponding to the shortest distance between the apical boundary and the costophrenic edge in the right chest image; calculating a second costophrenic edge point and a diaphragmatic edge point corresponding to the shortest distance between the costophrenic boundary and the transverse mediastinal edge in the right chest image; and obtaining the right lung segmentation image based on the first right lung apical edge point, the right lung longitudinal mediastinal edge point, the second right lung apical edge point, the first right lung costophrenic edge point, the second costophrenic edge point and the diaphragmatic edge point.

[0134] In the embodiments of the present disclosure and other possible embodiments, a DR image to be processed (a two-dimensional DR left lung image at multiple moments during the breathing process and / or in a deep inspiration state and / or in a deep expiration state) is obtained, and it is determined whether the DR image to be processed is a lung image according to the DR image to be processed; wherein the lung image is configured as a two-dimensional DR lung image at multiple moments during the breathing process; if the DR image to be processed is the lung image, a chest cavity detection is performed on the DR image to be processed, and information outside the chest cavity in the DR image to be processed is removed.

[0135] In the embodiments of the present disclosure and other possible embodiments, the method for determining whether the DR image to be processed is a lung image comprises: calculating an average gray value corresponding to the DR image to be processed, and determining whether the DR image to be processed is a lung image according to the average gray value and a set gray value. In the embodiments of the present disclosure and other possible embodiments, the set gray value can be configured by a person skilled in the art according to actual needs. For example, the set gray value can be configured as any one value or range between -1000HU and 0HU. Meanwhile, in the embodiments of the present disclosure and other possible embodiments, the DR image to be processed can also be determined whether it is a lung image by manual judgment.

[0136] In embodiments of the present disclosure and other possible embodiments, the method for determining whether the DR image is a lung image according to the average gray value and the set gray value comprises: if the to-be-processed DR image is not subjected to the inverting processing, if the average gray value is less than or equal to the set gray value, it is determined that the to-be-processed DR image is a lung image; if the to-be-processed DR image is subjected to the inverting processing, if the average gray value is greater than or equal to the set gray value, it is determined that the to-be-processed DR image is a lung image. The principle is that if it is a lung image, the lung in the lung image is generally full of air or contains a certain amount of air, and the air corresponds to a small gray value (CT value), which is generally configured as -1000 HU; and the gray value (CT value) of water is generally configured as 0 HU, and the gray value (CT value) of bone is generally configured as 1000 HU or more; therefore, the above technical solution is adopted to determine whether the to-be-processed DR image is a lung image.

[0137] In embodiments of the present disclosure and other possible embodiments, if it is the lung image, the to-be-processed DR image is subjected to the chest detection, and the information outside the chest in the to-be-processed DR image is removed. In embodiments of the present disclosure and other possible embodiments, the information outside the chest comprises: removing the unnecessary information such as arms, heads, and blank backgrounds.

[0138] In embodiments of the present disclosure, before the chest detection is performed on the to-be-processed DR image (the lung image to be segmented or the two-dimensional DR left lung image at multiple moments in the breathing process and / or in the deep inhalation state and / or in the deep exhalation state), the to-be-processed DR image is subjected to the filtering, and the filtered lung image to be segmented is subjected to the down-sampling to a set size.

[0139] In embodiments of the present disclosure, the logarithmic transformation of the to-be-processed DR image of the set size is subjected to the image enhancement, and an enhanced to-be-processed DR image is obtained.

[0140] (1) Image preprocessing of the to-be-processed DR image (the lung image to be segmented or the two-dimensional DR left lung image at multiple moments in the breathing process and / or in the deep inhalation state and / or in the deep exhalation state).

[0141] In embodiments and other possible embodiments of the present disclosure, a. the DR image to be processed (the lung image to be segmented) is processed, and low-pass filtering is performed on the DR image to be processed (the lung image to be segmented). The filtered lung image to be segmented is reduced (down-sampled) to a set size to speed up the processing speed of the image. Logarithmic transformation is performed on the down-sampled lung image to be segmented for image enhancement, and an enhanced lung to be segmented is obtained. The low-pass template Gaussian filter or mean filter of the set size value can be configured as 3x3 or 5x5. The range of the reduction (down-sampling) can be configured as 2-6 times. Those skilled in the art can configure the set size value and / or the range of the reduction (down-sampling) according to actual needs.

[0142] In embodiments and other possible embodiments of the present disclosure, b. the technical solution of the chest detection of the DR image to be processed adopts self-adaptive judgment of the chest contour according to the characteristics of the lung image to be segmented, and removes unnecessary information such as arm head and blank background.

[0143] In embodiments of the present disclosure, step 1). The method for processing the DR image to be processed includes: calculating a plurality of first gradient amplitudes corresponding to the horizontal direction of each pixel point in the DR image to be processed and a plurality of second gradient amplitudes of the vertical direction of each pixel point; determining a plurality of total gradient amplitudes based on the plurality of first gradient amplitudes and the plurality of second gradient amplitudes; integrating the plurality of first gradient amplitudes corresponding to the horizontal direction and the plurality of second gradient amplitudes of the vertical direction along the direction perpendicular to the direction, to obtain a plurality of first integral values and a plurality of second integral values; integrating the plurality of total gradient amplitudes in the vertical direction and the horizontal direction corresponding to the two directions, to obtain a plurality of third integral values and a plurality of fourth integral values; calculating a plurality of first local maximum values corresponding to the plurality of first ratios, and calculating a plurality of first local minimum values and a plurality of second local minimum values corresponding to the plurality of first ratios and the plurality of second ratios; determining the chest contour graph corresponding to the DR image to be processed according to the plurality of first local maximum values, the plurality of first local minimum values, the plurality of second local minimum values, and the chest contour characteristics. The chest contour characteristics can be configured as the first segmentation position of the neck or shoulder corresponding to the chest contour and the second segmentation position on both sides of the chest contour.

[0144] In embodiments of the present disclosure, the method for calculating a plurality of first gradient amplitudes corresponding to the horizontal direction of each pixel point in the DR image to be processed and a plurality of second gradient amplitudes of the vertical direction of each pixel point includes: obtaining a gradient operator; using the gradient operator, calculating a plurality of first gradient amplitudes corresponding to the horizontal direction of each pixel point in the DR image to be processed and a plurality of second gradient amplitudes of the vertical direction of each pixel point.

[0145] In embodiments of the present disclosure, the method for determining the plurality of total gradient magnitudes based on the plurality of first gradient magnitudes and the plurality of second gradient magnitudes comprises: calculating a plurality of first sum of squares corresponding to the plurality of first gradient magnitudes and a plurality of second sum of squares corresponding to the plurality of second gradient magnitudes, respectively, and determining the plurality of total gradient magnitudes based on the plurality of first sum of squares and the plurality of second sum of squares; and / or the method for determining the plurality of total gradient magnitudes based on the plurality of first sum of squares and the plurality of second sum of squares comprises: summing the plurality of first sum of squares and the plurality of second sum of squares, respectively, and square rooting the sum to obtain the plurality of total gradient magnitudes.

[0146] In embodiments of the present disclosure and other possible embodiments, the plurality of first gradient magnitudes corresponding to the horizontal direction of each pixel point in the lung image to be segmented and the plurality of second gradient magnitudes corresponding to the vertical direction of each pixel point are calculated, respectively; and the plurality of total gradient magnitudes are determined based on the plurality of first gradient magnitudes and the plurality of second gradient magnitudes; wherein the method for determining the plurality of total gradient magnitudes based on the plurality of first gradient magnitudes and the plurality of second gradient magnitudes comprises: calculating a plurality of first sum of squares corresponding to the plurality of first gradient magnitudes and a plurality of second sum of squares corresponding to the plurality of second gradient magnitudes, respectively, and determining the plurality of total gradient magnitudes based on the plurality of first sum of squares and the plurality of second sum of squares. Wherein the method for determining the plurality of total gradient magnitudes based on the plurality of first sum of squares and the plurality of second sum of squares comprises: summing the plurality of first sum of squares and the plurality of second sum of squares, respectively, and square rooting the sum to obtain the plurality of total gradient magnitudes.

[0147] For example, the matrix of each pixel point e in the lung image to be segmented and its eight adjacent areas is Using the sobel gradient operator The plurality of first gradient magnitudes (c+2*f+i-a-2*d-g) corresponding to the horizontal direction of each pixel point e in the lung image to be segmented and the transpose of the sobel gradient operator are calculated, respectively. The plurality of second gradient magnitudes (g+2*h+i-a-2*b-c) corresponding to the vertical direction of each pixel point e in the lung image to be segmented are calculated, respectively. Meanwhile, other gradient operators such as Roberts gradient operator or Laplace gradient operator can also be selected as needed by those skilled in the art.

[0148] For another example, the plurality of total gradient magnitudes are determined based on the plurality of first gradient magnitudes (h1, h2,..., h n ) and the plurality of second gradient magnitudes (k1, k2,..., k n ).

[0149] Step 2). Integrating the plurality of first gradient magnitudes corresponding to the horizontal direction and the plurality of second gradient magnitudes corresponding to the vertical direction along the direction perpendicular thereto to obtain a plurality of first integral values and a plurality of second integral values.

[0150] wherein the plurality of first gradient magnitudes corresponding to the horizontal direction are integrated in the vertical direction to obtain the plurality of first integral values, and the plurality of first gradient magnitudes corresponding to the vertical direction are integrated in the horizontal direction to obtain the plurality of second integral values.

[0151] Step 3). Integrating the plurality of total gradient magnitudes in the vertical direction and the horizontal direction corresponding thereto to obtain a plurality of third integral values and a plurality of fourth integral values; wherein the plurality of total gradient magnitudes are integrated in the vertical direction to obtain the plurality of third integral values, and the plurality of total gradient magnitudes are integrated in the horizontal direction to obtain the plurality of fourth integral values.

[0152] Step 4). (a) When a plurality of first local maximum values are determined, a ratio is calculated according to the results of Step 2 and Step 3, and when the ratio is less than a preset value, the image information at this position is regarded as noise and needs to be discarded. Wherein the noise can be configured as image information corresponding to unnecessary information such as arm head and blank background.

[0153] wherein, based on the plurality of first integral values and the plurality of third integral values, it is determined whether each of the plurality of first ratios corresponding to the plurality of first integral values and the plurality of third integral values in the horizontal direction is retained.

[0154] wherein, based on the plurality of first integral values and the plurality of third integral values, it is determined whether each of the plurality of first ratios corresponding to the plurality of first integral values and the plurality of third integral values in the horizontal direction is retained. The method comprises: obtaining a first preset value; calculating a plurality of first ratios of the plurality of first integral values and the plurality of third integral values corresponding thereto in the horizontal direction; and discarding a certain first ratio in the plurality of first ratios if the certain first ratio is less than the first preset value. In the embodiments of the present disclosure and other possible embodiments, the first preset value can also need to be configured according to actual needs for those skilled in the art.

[0155] (b) When a plurality of first local minimum values are determined, a ratio is calculated according to the results of Step 2 and Step 3, and when the ratio is greater than a preset value, the image information at this position is regarded as noise and needs to be discarded. Wherein the noise can be configured as image information corresponding to unnecessary information such as arm head and blank background.

[0156] wherein, based on the plurality of first integral values and the plurality of third integral values, it is determined whether each of a plurality of first ratios corresponding to the plurality of first integral values and the plurality of third integral values in the horizontal direction is retained.

[0157] wherein, based on the plurality of first integral values and the plurality of third integral values, it is determined whether each of a plurality of first ratios corresponding to the plurality of first integral values and the plurality of third integral values in the horizontal direction is retained.

[0158] wherein, based on the plurality of second integral values and the plurality of fourth integral values, it is determined whether each of a plurality of second ratios corresponding to the plurality of second integral values and the plurality of fourth integral values in the vertical direction is retained.

[0159] wherein, based on the plurality of second integral values and the plurality of fourth integral values, it is determined whether each of a plurality of second ratios corresponding to the plurality of second integral values and the plurality of fourth integral values in the vertical direction is retained.

[0160] In the embodiments of the present disclosure, before the plurality of first local maximum values corresponding to the plurality of first ratios are calculated, and the plurality of first local minimum values and the plurality of second local minimum values corresponding to the plurality of first ratios and the plurality of second ratios are calculated, based on the plurality of first integral values and the plurality of third integral values, it is determined whether each of a plurality of first ratios corresponding to the plurality of first integral values and the plurality of third integral values in the horizontal direction is retained; and, based on the plurality of second integral values and the plurality of fourth integral values, it is determined whether each of a plurality of second ratios corresponding to the plurality of second integral values and the plurality of fourth integral values in the vertical direction is retained; further, the plurality of first local maximum values corresponding to the plurality of retained first ratios are calculated, and the plurality of first local minimum values and the plurality of second local minimum values corresponding to the plurality of retained first ratios and the plurality of retained second ratios are calculated.

[0161] In embodiments of the present disclosure, the differential derivatives of the first curve corresponding to the plurality of first ratios are calculated to obtain the plurality of first local maximum values and the plurality of first local minimum values; and the differential derivatives of the second curve corresponding to the plurality of second ratios are calculated to obtain the plurality of second local minimum values.

[0162] In embodiments of the present disclosure, the method for determining the thoracic profile corresponding to the lung image to be segmented according to the plurality of first local maximum values, the plurality of first local minimum values, the plurality of second local minimum values, and the thoracic feature includes: determining the second segmentation positions on both sides of the thoracic profile based on the plurality of first local maximum values and the thoracic feature; determining the first segmentation positions of the neck or shoulder based on the plurality of second local minimum values and the thoracic feature; and determining the thoracic profile corresponding to the lung image to be segmented according to the first segmentation positions and the first segmentation positions.

[0163] In embodiments of the present disclosure and other possible embodiments, 5). The plurality of first local maximum values and the plurality of first local minimum values corresponding to the plurality of first ratios after the discard of step 4(a) are calculated respectively. Similarly, the plurality of first local minimum values and the plurality of second local minimum values corresponding to the plurality of first ratios and the plurality of second ratios after the discard of step 4(b) are calculated respectively. Wherein, the plurality of first ratios after the discard are the plurality of first ratios retained after the discard; similarly, the plurality of second ratios after the discard are the plurality of second ratios retained after the discard.

[0164] According to the plurality of first local maximum values, the plurality of first local minimum values, the plurality of second local minimum values, and the thoracic feature, the thoracic profile corresponding to the lung image to be segmented is determined, and unnecessary information such as arms, heads, and blank backgrounds is removed. Wherein, the thoracic feature can configure the first segmentation positions of the neck or shoulder corresponding to the thoracic profile and the second segmentation positions on both sides of the thoracic profile.

[0165] Wherein, the method for determining the thoracic profile corresponding to the lung image to be segmented according to the plurality of first local maximum values, the plurality of first local minimum values, the plurality of second local minimum values, and the thoracic feature includes: determining the second segmentation positions on both sides of the thoracic profile based on the plurality of first local maximum values and the thoracic feature; determining the first segmentation positions of the neck or shoulder based on the plurality of second local minimum values and the thoracic feature; and determining the thoracic profile corresponding to the lung image to be segmented according to the first segmentation positions and the first segmentation positions.

[0166] The first local maximum values are obtained by calculating the differential derivative of the first curve corresponding to the discarded first ratios, and the first local minimum values are obtained by calculating the differential derivative of the first curve corresponding to the discarded first ratios. The differential derivative can be configured as a first-order, second-order or other multi-order differential derivative. Similarly, the second local minimum values are obtained by calculating the differential derivative of the second curve corresponding to the discarded second ratios, and the differential derivative can be configured as a first-order, second-order or other multi-order differential derivative.

[0167] In an embodiment of the present disclosure, the method for determining the second segmentation position of the two sides of the thoracic cavity based on the plurality of first local maximum values and the thoracic cavity features comprises: determining the maximum value of all the plurality of first local maximum values on one side of the center line of the to-be-processed DR image, and configuring the position information corresponding to the maximum value as the to-be-determined one-side segmentation position point corresponding to the one side of the thoracic cavity; determining the minimum value of all the plurality of first local minimum values on the other side of the center line of the to-be-processed DR image, and configuring the position information corresponding to the minimum value as the to-be-determined other-side segmentation position point corresponding to the other side of the thoracic cavity.

[0168] In an embodiment of the present disclosure, the method for determining the first segmentation position of the neck or shoulder based on the plurality of second local minimum values and the thoracic cavity features comprises: configuring the position information corresponding to the minimum value of the plurality of second local minimum values as the first segmentation position of the neck or shoulder.

[0169] In an embodiment of the present disclosure and other possible embodiments, the method for determining the second segmentation position of the two sides of the thoracic cavity based on the plurality of first local maximum values and the thoracic cavity features comprises: determining the maximum value of all the plurality of first local maximum values on one side (right side) of the center line of the to-be-processed DR image, and configuring the position information corresponding to the maximum value as the to-be-determined one-side segmentation position point corresponding to the one side of the thoracic cavity; determining the minimum value of all the plurality of first local minimum values on the other side (left side) of the center line of the to-be-processed DR image, and configuring the position information corresponding to the minimum value as the to-be-determined other-side segmentation position point corresponding to the other side of the thoracic cavity.

[0170] For example, the maximum value a n ) of all the plurality of first local maximum values on one side (right side) of the center line is determined, and the position information corresponding to the maximum value a m; wherein, m is less than or equal to n; m is the position information (for example, a certain horizontal coordinate) corresponding to the maximum value, that is, the side corresponding to the thoracic side of the to-be-determined side segmentation position point. Similarly, the minimum value b n of all the first local minimum values (b1, b2,..., b r on the other side (left side) of the center line is determined, wherein r is less than or equal to n; r is the position information (for example, a certain horizontal coordinate) corresponding to the minimum value.

[0171] In the embodiments and other possible embodiments of the present disclosure, the method for determining the first segmentation position of the neck or shoulder based on the plurality of second local minimum values and the thoracic characteristics includes: configuring the position information (for example, a certain longitudinal coordinate) corresponding to the minimum value corresponding to the plurality of second local minimum values as the first segmentation position of the neck or shoulder.

[0172] In the embodiments of the present disclosure, the thoracic profile obtained by performing thoracic cavity detection on the to-be-processed DR image is configured as a to-be-segmented lung image; and based on the to-be-segmented lung image, segmentation of the left lung and the right lung is performed. Wherein, the to-be-segmented lung image is configured as a plurality of DR lung images at multiple moments in the breathing process and / or in the deep inspiration state and / or in the deep expiration state.

[0173] In the embodiments of the present disclosure, the to-be-segmented lung image is segmented into a left thoracic image and a right thoracic image; based on the left thoracic image and the right thoracic image respectively, segmentation of the left lung and the right lung is performed. Wherein, the to-be-segmented lung image is configured as a plurality of DR lung images at multiple moments in the breathing process and / or in the deep inspiration state and / or in the deep expiration state.

[0174] In the embodiments of the present disclosure, the method for respectively extracting the rib border in the left lung image and / or the rib border in the right lung image corresponding to the plurality of DR lung images in the deep inspiration state and the deep expiration state includes: performing left lung and right lung segmentation on the plurality of DR lung images in the deep inspiration state and the deep expiration state respectively, to obtain the left lung image and / or the right lung image corresponding to the plurality of DR lung images; performing rib border detection on the left lung image and / or the right lung image respectively, to obtain the rib border in the left lung image and / or the rib border in the right lung image corresponding to the plurality of DR lung images.

[0175] In embodiments of the present disclosure, the method for detecting the costophrenic boundary in the left lung image and / or the right lung image respectively, to obtain the costophrenic boundary in the left lung image and / or the right lung image corresponding to the plurality of DR lung images, comprises: extracting a plurality of left lung contour lines and / or a plurality of right lung contour lines corresponding to the left lung image and / or the right lung image corresponding to the plurality of DR lung images respectively; and determining the costophrenic boundary in the left lung image and / or the right lung image corresponding to the plurality of DR lung images according to the plurality of left lung contour lines and / or the plurality of right lung contour lines respectively.

[0176] In embodiments of the present disclosure, the method for determining the costophrenic boundary in the left lung image and / or the right lung image corresponding to the plurality of DR lung images according to the plurality of left lung contour lines and the plurality of right lung contour lines respectively, comprises: determining a plurality of left lung motion peaks, a plurality of left lung motion troughs, and / or a plurality of right lung motion peaks, and a plurality of right lung motion troughs according to the plurality of left lung contour lines and / or the plurality of right lung contour lines respectively; determining the costophrenic boundary in the left lung image corresponding to the plurality of DR lung images based on the plurality of left lung motion peaks and the plurality of left lung motion troughs; and / or determining the costophrenic boundary in the right lung image corresponding to the plurality of DR lung images based on the plurality of right lung motion peaks and the plurality of right lung motion troughs.

[0177] In embodiments of the present disclosure, the method for determining the costophrenic boundary in the left lung image corresponding to the plurality of DR lung images based on the plurality of left lung motion peaks and the plurality of left lung motion troughs, comprises: connecting the plurality of left lung motion peaks and the plurality of left lung motion troughs corresponding thereto to obtain a plurality of first division lines; and configuring the costophrenic boundary of the left lung image on the right side of the plurality of first division lines as the costophrenic boundary in the left lung image corresponding to the plurality of DR lung images. The plurality of first division lines obtained by connecting the plurality of left lung motion peaks and the plurality of left lung motion troughs corresponding thereto are respectively the boundaries of the plurality of left lung contour lines close to the left costophrenic boundary.

[0178] In embodiments of the present disclosure, the method for determining the costophrenic boundary in the right lung image corresponding to the plurality of DR lung images based on the plurality of right lung motion peaks and the plurality of right lung motion troughs, comprises: connecting the plurality of right lung motion peaks and the plurality of right lung motion troughs corresponding thereto to obtain a plurality of second division lines; and configuring the costophrenic boundary of the right lung image on the left side of the plurality of second division lines as the costophrenic boundary in the right lung image corresponding to the plurality of DR lung images. The plurality of second division lines obtained by connecting the plurality of right lung motion peaks and the plurality of right lung motion troughs corresponding thereto are respectively the boundaries of the plurality of right lung contour lines close to the right costophrenic boundary.

[0179] In the embodiments of the present disclosure, the method for performing left lung and right lung segmentation on the plurality of DR lung images in the deep inhalation state and the deep exhalation state respectively to obtain the left lung images and the right lung images corresponding to the plurality of DR lung images comprises: performing rib boundary, lung apex boundary and mediastinum and diaphragm edge detection on the left chest images and the right chest images of the plurality of DR lung images in the deep inhalation state and the deep exhalation state respectively to obtain the left lung images and the right lung images corresponding to the plurality of DR lung images. Alternatively, the method for performing left lung and right lung segmentation on the plurality of DR lung images in the deep inhalation state and the deep exhalation state respectively to obtain the left lung images and the right lung images corresponding to the plurality of DR lung images comprises: obtaining a segmentation model of a preset convolutional neural network and a DR lung region label image used for training the segmentation model; training the segmentation model by using the DR lung region label image used for training the segmentation model; and performing left lung and right lung segmentation on the plurality of DR lung images in the deep inhalation state and the deep exhalation state based on the trained segmentation model to obtain the left lung images and the right lung images corresponding to the plurality of DR lung images.

[0180] In the embodiments and other possible embodiments of the present disclosure, the segmentation model of the preset convolutional neural network is configured as a Unet convolutional neural network or a nnUnet convolutional neural network or a convolutional neural network improved based on the Unet convolutional neural network or a convolutional neural network improved based on the nnUnet convolutional neural network. For example, the convolutional neural network improved based on the Unet convolutional neural network can be configured as a ResUnet convolutional neural network with a residual structure.

[0181] In the embodiments and other possible embodiments of the present disclosure, the Unet convolutional neural network or the nnUnet convolutional neural network or the convolutional neural network improved based on the Unet convolutional neural network or the convolutional neural network improved based on the nnUnet convolutional neural network at least comprises: a down-sampling contraction path, an up-sampling expansion path and a final classification layer.

[0182] In the embodiments and other possible embodiments of the present disclosure, the plurality of DR lung region images are configured as a plurality of DR lung region images collected at multiple time points in the deep inhalation state or the breath-hold state.

[0183] c. According to the thoracic cage image corresponding to the lung image to be segmented, the thoracic cage image is segmented into a left chest image and a right chest image, and based on this, the left lung field of the left chest image is segmented and the right lung field of the right chest image is segmented.

[0184] In the embodiments of the present disclosure, the method for left lung and right lung segmentation based on the left chest image and the right chest image respectively comprises: rib edge boundary, lung apex boundary and mediastinum and diaphragm edge detection of the left chest image and the right chest image respectively; obtaining a left lung segmentation image according to the rib edge boundary, lung apex boundary and mediastinum and diaphragm edge of the left chest image; obtaining a right lung segmentation image according to the rib edge boundary, lung apex boundary and mediastinum and diaphragm edge of the right chest image.

[0185] In the embodiments of the present disclosure, the method for rib edge boundary of the left chest image comprises: constructing a direction derivative template of the left chest image by using a direction derivative, and setting a set weighting depth of the direction derivative template; performing template traversal of the direction derivative on the left chest image by using the direction derivative template corresponding to the set weighting depth, superimposing the result of the template traversal into the left chest image to obtain a left chest superimposed image; performing binaryzation processing on the left chest superimposed image to obtain a left rib edge binary image; obtaining a left rib edge angle image to be screened according to the left rib edge binary image and the left chest superimposed image; obtaining a screened left rib edge angle image based on the left rib edge angle image to be screened and a first set rib edge angle; performing connected domain selection on the left rib edge angle image to obtain a rib edge boundary corresponding to a maximum connected domain.

[0186] In the embodiments of the present disclosure, the method for obtaining a left rib edge angle image to be screened according to the left rib edge binary image and the left chest superimposed image comprises: performing morphological opening and closing operation and thinning processing on the left rib edge binary image to obtain a morphologically processed left rib edge binary image; performing AND operation on the gradient direction angle of each pixel in the morphologically processed left rib edge binary image and the left chest superimposed image to obtain a left rib edge angle image to be screened.

[0187] In the embodiments of the present disclosure, before the direction derivative template of the left chest image is constructed by using the direction derivative, a set scale Gaussian blur is performed on the left chest image to obtain a corresponding left chest Gaussian blur image; then, the direction derivative template of the left chest Gaussian blur image is constructed by using the direction derivative; in the process of rib edge boundary of the left chest image, the direction derivative template traversal is performed on the left chest Gaussian blur image by using the direction derivative template corresponding to the set weighting depth, the result of the template traversal is superimposed into the left chest image to obtain a left chest Gaussian blur superimposed image; the binaryzation processing is performed on the left chest Gaussian blur superimposed image to obtain a left rib edge binary image; the left rib edge angle image to be screened is obtained according to the left rib edge binary image and the left chest Gaussian blur superimposed image.

[0188] a.In embodiments of the present disclosure and other possible embodiments, the lung field segmentation step is as follows, taking the left chest image as an example. For example, the rib edge boundary detection of the left chest image.

[0189] 1). The left chest region (image) is set to a scale Gaussian blur, the details of the left chest region (image) are reduced, and a corresponding processed left chest Gaussian blur image is obtained. Wherein, the set scale can be configured as 7x7 or 9x9, and the skilled in the art can configure the set scale according to actual needs. Wherein, the mean square deviation of the Gaussian blur algorithm can be configured as 2, 2.5, 3 and the like, and the skilled in the art can configure the mean square deviation σ of the Gaussian blur algorithm according to actual needs.

[0190] 2). A directional derivative template of the processed left chest Gaussian blur image f(x0, y0) is constructed using the directional derivative, and a set weighting depth of the directional derivative template is set. Wherein, (x0, y0) are the horizontal coordinate x0 and the vertical coordinate y0 in the corresponding coordinate point of the left chest image f(x0, y0).

[0191] Wherein, the calculation formula of the directional derivative is as follows:

[0192]

[0193] Wherein, l is the unit vector in the direction, and cosα and cosβ are the cosine of the direction. Wherein, the direction can be configured as the horizontal direction, α is the angle formed by the l direction and the horizontal direction, and β is the angle formed by the l direction and the vertical direction.

[0194] Wherein, the method of constructing the directional derivative template of the processed left chest Gaussian blur image f(x0, y0) using the directional derivative includes: obtaining a first radius in the x0 direction and a second radius in the y0 direction corresponding to the set template radius r. Based on the first radius and the second radius , the directional derivative template of the processed left chest Gaussian blur image f(x0, y0) is constructed using the directional derivative.

[0195]

[0196] Wherein,

[0197]

[0198] For example, the set template radius can be configured as 1, 2, 3, 4, 5 and the like. The skilled in the art can configure the set template radius according to actual needs.

[0199] For example, the first radius in the x0 direction is 1 The second radius in the y0 direction is configured as -1 or 0 or 1 The value range of the second radius in the y0 direction is -1, 0 and 1.

[0200] Wherein, the setting weight depth can be configured by the person skilled in the art according to actual needs, for example, the setting weight depth can be configured as 6. In addition, the method for setting the setting weight depth of the direction derivative template comprises: obtaining the setting weight depth, multiplying the setting weight depth by the direction derivative template to obtain the direction derivative template corresponding to the setting weight depth. According to the structural characteristics of the costophrenic region, the direction angle range is reasonably set and brought into the direction derivative calculation formula to obtain the template array.

[0201] 3). Using the direction derivative template corresponding to the setting weight depth, the left chest Gaussian blur image processed in step a.1 is subjected to template traversal of the direction derivative, and the result of the template traversal is superimposed (the corresponding pixels are added) into the left chest Gaussian blur image in step 1 to obtain a left chest Gaussian blur superimposed image.

[0202] For example, the left chest Gaussian blur image each pixel point e and its eight neighborhood matrix are Using the direction derivative template corresponding to the setting weight depth The superimposed value (e+w*(c+2*f+i-a-2*d-g)) of each pixel point e is calculated respectively.

[0203] 4). The result (left chest Gaussian blur superimposed image) of step a.3 is subjected to binaryzation processing of maximum inter-class variance to obtain a costophrenic binarization image.

[0204] 5). The result (costophrenic binarization image) of step a.4 is subjected to morphological opening and closing operation and thinning processing to obtain a morphologically processed costophrenic binarization image.

[0205] 6). The result (left chest Gaussian blur superimposed image) of step a.3 is subjected to horizontal gradient and vertical gradient calculation respectively to obtain a horizontal gradient image and a vertical gradient image; based on the horizontal gradient image and the vertical gradient image, the gradient direction angle of each pixel in the left chest Gaussian blur superimposed image is calculated.

[0206] 7). The result (morphologically processed costophrenic binarization image) of step a.5 is subjected to AND operation with the result (gradient direction angle of each pixel in the left chest Gaussian blur superimposed image) of step a.6 to obtain a to-be-screened costophrenic angle image.

[0207] 8). Set a reasonable angle range according to the characteristics of the rib edge tissue (first set rib edge angle range), and perform angle screening on the result obtained in step a.7 (the rib edge angle graph to be screened), to obtain a screened rib edge angle graph.

[0208] 9). According to the result in step a.8 (the screened rib edge angle graph), perform connected domain selection to obtain the maximum connected domain, and the connected domain is the rib edge part (rib edge boundary).

[0209] Similarly, in the embodiments of the present disclosure, the method for performing rib edge boundary on the right chest image comprises: constructing a direction derivative template of the right chest image by using a direction derivative, and setting a set weighting depth of the direction derivative template; performing template traversal of the direction derivative on the right chest image by using the direction derivative template corresponding to the set weighting depth, and superimposing the result of the template traversal into the right chest image to obtain a right chest superimposed image; performing binaryzation processing on the right chest superimposed image to obtain a right rib edge binary graph; obtaining a right rib edge angle graph to be screened according to the right rib edge binary graph and the right chest superimposed image; obtaining a screened right rib edge angle graph based on the right rib edge angle graph to be screened and a second set rib edge angle; and performing connected domain selection on the right rib edge angle graph to obtain a rib edge boundary corresponding to the maximum connected domain.

[0210] Similarly, in the embodiments of the present disclosure, the method for obtaining the right rib edge angle graph to be screened according to the right rib edge binary graph and the right chest superimposed image comprises: performing morphological opening and closing operation and thinning processing on the right rib edge binary graph to obtain a morphologically processed right rib edge binary graph; and performing AND operation on the gradient direction angle of each pixel in the morphologically processed right rib edge binary graph and the right chest superimposed image to obtain the right rib edge angle graph to be screened.

[0211] Similarly, in the embodiments of the present disclosure, before the direction derivative template of the right chest image is constructed by using the direction derivative, a right chest Gaussian blur image corresponding to the right chest image is obtained by performing a set scale Gaussian blur on the right chest image; then, the direction derivative template of the right chest Gaussian blur image is constructed by using the direction derivative; in the process of performing rib edge boundary on the right chest image, the direction derivative template corresponding to the set weighting depth is used to perform template traversal of the direction derivative on the right chest Gaussian blur image, and the result of the template traversal is superimposed into the right chest image to obtain a right chest Gaussian blur superimposed image; the right rib edge binary graph is obtained by performing binaryzation processing on the right chest Gaussian blur superimposed image; and the right rib edge angle graph to be screened is obtained according to the right rib edge binary graph and the right chest Gaussian blur superimposed image.

[0212] b. Lung apex boundary detection. In embodiments of the present disclosure, the method of performing left lung apex boundary detection on the left chest image comprises: determining a left lung apex detection region according to the left chest image; determining a left lung apex edge binary image according to the left lung apex detection region; and obtaining a fitted left lung apex boundary by using a quadratic function fitting according to the left lung apex edge binary image.

[0213] In embodiments of the present disclosure, the method of determining a left lung apex detection region according to the left chest image comprises: detecting a first coordinate corresponding to the uppermost coordinate point of the rib margin of the left chest image; and configuring a region with a hypotenuse formed by the first coordinate and the uppermost coordinate point of the rightmost corner of the left chest image as the left lung apex detection region.

[0214] In embodiments of the present disclosure and other possible embodiments, 1). The coordinate corresponding to the uppermost coordinate point of the rib margin of the detected left chest image. 2). The coordinate point and the uppermost coordinate point of the leftmost corner of the left chest image form a rectangular region with a hypotenuse as the lung apex detection region. For the right lung corresponding lung apex detection region, the coordinate corresponding to the uppermost coordinate point of the rib margin and the uppermost coordinate point of the leftmost corner of the right chest image form a rectangular region with a hypotenuse as the right lung corresponding lung apex detection region. 3). Perform scale Gaussian filtering and contrast enhancement on the lung apex region obtained in step b.2 to obtain a filtered and enhanced lung apex region image. 4). Based on the filtered and enhanced lung apex region image, the same method as steps a.2 to a.5 is used to determine the left lung apex edge binary image according to the angle characteristics of the lung apex edge. Wherein, the angle characteristics of the lung apex edge are configured to set an angle range. 5). According to the characteristics of the lung apex edge and the left lung apex edge binary image, a quadratic function Hough space parameter fitting is used to obtain a fitted lung apex edge (line). Wherein, the characteristics of the lung apex edge are a quadratic function with an opening downward.

[0215] Similarly, in embodiments of the present disclosure, the method of performing right lung apex boundary detection on the right chest image comprises: determining a right lung apex detection region according to the right chest image; determining a right lung apex edge binary image according to the right lung apex detection region; and obtaining a fitted right lung apex boundary by using a quadratic function fitting according to the right lung apex edge binary image.

[0216] Similarly, in embodiments of the present disclosure, the method of determining a right lung apex detection region according to the right chest image comprises: detecting a second coordinate corresponding to the uppermost coordinate point of the rib margin of the right chest image; and configuring a region with a hypotenuse formed by the second coordinate and the uppermost coordinate point of the leftmost corner of the left chest image as the right lung apex detection region.

[0217] c. mediastinal and diaphragmatic edge detection. In embodiments of the present disclosure, a method for left lung mediastinal and diaphragmatic edge detection on the left chest image respectively, comprises: performing binaryzation on the left chest image to obtain a left chest binary image; performing edge detection on the left chest binary image to obtain a left chest edge binary image; obtaining a left chest edge angle image according to the gradient direction angle of each pixel in the left chest binary image and the left chest edge binary image; obtaining a selected left diaphragmatic and mediastinal edge angle image according to the obtained left chest edge angle image and the edge angle range of the set diaphragm and mediastinum; performing connected domain selection processing according to the selected left diaphragmatic and mediastinal edge angle image to obtain the left lung mediastinal and diaphragmatic edge corresponding to the maximum connected domain.

[0218] In embodiments of the present disclosure, the method for performing binaryzation on the left chest image to obtain a left chest binary image comprises: performing contrast enhancement processing and maximum inter-class variance processing on the left chest Gaussian blur image corresponding to the left chest image to obtain a left chest binary image.

[0219] In embodiments of the present disclosure, the method for determining the gradient direction angle of each pixel in the left chest edge binary image comprises: performing horizontal gradient and vertical gradient calculation on the left chest Gaussian blur image corresponding to the left chest image to obtain a horizontal gradient image and a vertical gradient image of the left chest; and obtaining the gradient direction angle of each pixel in the left chest Gaussian blur image based on the horizontal gradient image and the vertical gradient image of the left chest.

[0220] In the embodiments and other possible embodiments of the present disclosure, the left lung mediastinum and transverse septum edge detection method for the left chest image includes: 1) performing contrast enhancement processing and maximum inter-class variance processing on the processed left chest Gaussian blur image obtained in step a.1 to obtain a left chest binary image; 2) sequentially performing morphological opening and closing operations and canny edge detection on the result of step c.1 (left chest binary image) to obtain a left chest edge binary image; 3) performing horizontal gradient and vertical gradient calculation on the processed left chest Gaussian blur image obtained in step a.1 to obtain a horizontal gradient image and a vertical gradient image; based on the horizontal gradient image and the vertical gradient image, the gradient direction angle of each pixel in the left chest Gaussian blur image is calculated; 4) performing AND operation on the left chest binary image obtained in step c.1 and the left chest edge binary image obtained in c.2 to retain the angle on the edge pixel point, to obtain a left chest edge angle image; 5) selecting a suitable angle range according to the edge characteristics of the transverse septum and the mediastinum (setting the edge angle range of the transverse septum and the mediastinum), removing the stray tissue edge information in the left chest edge angle image, to obtain a selected left transverse septum and mediastinum edge angle image; 6) performing connected domain selection processing according to the result of c.5 (selected left transverse septum and mediastinum edge angle image) to obtain the maximum connected domain, which is the transverse septum and mediastinum edge region.

[0221] Similarly, in the embodiments of the present disclosure, the right lung mediastinum and transverse septum edge detection method for the right chest image respectively includes: performing binaryzation processing on the right chest image to obtain a right chest binary image; performing edge detection on the right chest binary image to obtain a right chest edge binary image; obtaining a right chest edge angle image according to the gradient direction angle of each pixel in the right chest binary image and the right chest edge binary image; obtaining a selected right transverse septum and mediastinum edge angle image according to the right chest edge angle image and the set edge angle range of the transverse septum and the mediastinum; and performing connected domain selection processing according to the selected right transverse septum and mediastinum edge angle image to obtain the right lung mediastinum and transverse septum edge corresponding to the maximum connected domain.

[0222] Similarly, in the embodiments of the present disclosure, the method of performing binaryzation processing on the right chest image to obtain a right chest binary image includes: performing contrast enhancement processing and maximum inter-class variance processing on the right chest Gaussian blur image corresponding to the right chest image to obtain a right chest binary image.

[0223] Similarly, in the embodiments of the present disclosure, the method for determining the gradient direction angle of each pixel in the right chest edge binary image comprises: performing transverse gradient and longitudinal gradient calculation on the right chest Gaussian blur image corresponding to the right chest image to obtain the transverse gradient image and the longitudinal gradient image of the right chest; and obtaining the gradient direction angle of each pixel in the right chest Gaussian blur image based on the transverse gradient image and the longitudinal gradient image of the right chest.

[0224] (3) The connection of the transverse diaphragm edge, the lung apex and the rib edge. In the embodiments of the present disclosure, the method for obtaining the left lung segmentation image according to the rib edge boundary, the lung apex boundary and the longitudinal and transverse diaphragm edges corresponding to the left chest image comprises: calculating a first left lung apex edge point and a left lung longitudinal diaphragm edge point corresponding to the shortest distance between the lung apex boundary and the longitudinal diaphragm edge in the left chest image; calculating a second left lung apex edge point and a first left lung rib edge point corresponding to the shortest distance between the lung apex boundary and the rib edge in the left chest image; calculating a second rib edge point and a transverse diaphragm edge point corresponding to the shortest distance between the rib edge boundary and the transverse diaphragm edge in the left chest image; and obtaining the left lung segmentation image based on the first left lung apex edge point, the left lung longitudinal diaphragm edge point, the second left lung apex edge point, the first left lung rib edge point, the second rib edge point and the transverse diaphragm edge point.

[0225] Similarly, in the embodiments of the present disclosure, the method for obtaining the right lung segmentation image according to the rib edge boundary, the lung apex boundary and the longitudinal and transverse diaphragm edges corresponding to the right chest image comprises: calculating a first right lung apex edge point and a right lung longitudinal diaphragm edge point corresponding to the shortest distance between the lung apex boundary and the longitudinal diaphragm edge in the right chest image; calculating a second right lung apex edge point and a first right lung rib edge point corresponding to the shortest distance between the lung apex boundary and the rib edge in the right chest image; calculating a second rib edge point and a transverse diaphragm edge point corresponding to the shortest distance between the rib edge boundary and the transverse diaphragm edge in the right chest image; and obtaining the right lung segmentation image based on the first right lung apex edge point, the right lung longitudinal diaphragm edge point, the second right lung apex edge point, the first right lung rib edge point, the second rib edge point and the transverse diaphragm edge point.

[0226] In embodiments of the present disclosure and other possible embodiments, the method for connecting the edges of the transverse diaphragm, the lung apex and the rib edge comprises: a. calculating the two points (the first lung apex edge point and the transverse diaphragm edge point) between which the Euclidean distance between the lung apex edge (line) and the transverse diaphragm edge (line) is the shortest, and the two points are respectively one of the end points of the lung apex and the transverse diaphragm boundary; b. calculating the two points (the second lung apex edge point and the first rib edge point) between which the Euclidean distance between the lung apex edge (line) and the rib edge (line) is the shortest, and the two points are respectively one of the end points of the lung apex and the rib edge, and the other lung apex end point obtained in step a is the lung apex boundary region; c. calculating the two points (the second rib edge point and the transverse diaphragm edge point) between which the Euclidean distance between the rib edge (line) and the transverse diaphragm edge (line) is the shortest, and the two points are respectively one of the end points of the rib edge and the transverse diaphragm, and the other transverse diaphragm boundary end point obtained in step a is the transverse diaphragm boundary region, and the other rib edge end point obtained in step b is the rib edge boundary region; d. connecting the three edge regions obtained in steps a, b and c to obtain a closed lung field contour, and according to the difference between the tissues inside and outside the contour boundary, the lung field region can be segmented; e. the right lung field region is processed in the same way as described above, and finally the lung field region is mapped to the original image to obtain the lung field segmentation of the original image.

[0227] Further, the segmentation model is trained using the DR lung region label image used for training the segmentation model. In embodiments of the present disclosure and other possible embodiments, before the segmentation model is trained using the DR lung region label image used for training the segmentation model, the DR lung region label image is data-augmented to obtain an augmented DR lung region label image; and the segmentation model is trained using the augmented DR lung region label image.

[0228] In embodiments of the present disclosure and other possible embodiments, the method for data-augmenting the DR lung region label image to obtain an augmented DR lung region label image comprises: performing spatial geometric transformation and / or flipping and / or rotation and / or cropping and / or scaling and / or image shifting and / or edge padding and / or random erasing and / or random occlusion operation on the DR lung region label image to obtain the augmented DR lung region label image.

[0229] In embodiments of the present disclosure and other possible embodiments, the method for data-augmenting the DR lung region label image to obtain an augmented DR lung region label image further comprises: randomly extracting any two DR lung region label images from the DR lung region label image; performing configuration operation on the any two DR lung region label images to obtain corresponding DR lung region label registration images; and performing fusion operation on the DR lung region label registration images to obtain the augmented DR lung region label image.

[0230] In embodiments of the present disclosure and other possible embodiments, the method for performing the fusion operation on the DR lung region label registration image to obtain an enhanced DR lung region label image comprises: performing a minimum value taking operation, a maximum value taking operation or a mean value taking operation on pixel values corresponding to the DR lung region label registration image respectively to obtain an enhanced DR lung region label image.

[0231] Finally, based on the trained segmentation model, left lung and / or right lung segmentation of the plurality of DR lung images to be segmented is completed.

[0232] Step S102: determining a first motion displacement according to a rib edge boundary in a left lung image corresponding to the plurality of DR lung images; and / or determining a second motion displacement according to a rib edge boundary in a right lung image corresponding to the plurality of DR lung images.

[0233] In embodiments of the present disclosure, the method for determining the first motion displacement according to the rib edge boundary in the left lung image corresponding to the plurality of DR lung images comprises: performing registration on the rib edge boundary in the left lung image corresponding to the plurality of DR lung images to obtain a left lung registration rib edge boundary; and determining the first motion displacement according to the left lung registration rib edge boundary.

[0234] In embodiments of the present disclosure, the method for determining the first motion displacement according to the left lung registration rib edge boundary comprises: calculating a plurality of first distances corresponding to each boundary point pair in the left lung registration rib edge boundary based on the left lung registration rib edge boundary; and configuring the plurality of first distances as the first motion displacement.

[0235] In embodiments of the present disclosure and other possible embodiments, the left lung registration rib edge boundary contains a one-to-one correspondence relationship (boundary point pair after registration) of each boundary point in the left lung registration rib edge boundary before registration. Therefore, the first distance of the position of each boundary point pair in the left lung registration rib edge boundary before registration and after registration can be calculated, and all first distances of the positions of all boundary point pairs are the plurality of first distances.

[0236] Or, in embodiments of the present disclosure, the method for determining the first motion displacement according to the left lung registration rib edge boundary comprises: based on a plurality of positions of the left lung rib or a plurality of set positions, dividing the left lung registration rib edge boundary before registration and after registration respectively to obtain a first plurality of segmented left lung rib edge boundaries before registration and a first plurality of segmented left lung rib edge boundaries after registration; and based on the left lung registration rib edge boundary, calculating a plurality of first distances corresponding to each boundary point in the first plurality of segmented left lung rib edge boundaries before registration and after registration respectively, and configuring the plurality of first distances as the first motion displacement.

[0237] In embodiments and other possible embodiments of the present disclosure, a person skilled in the art can configure the plurality of positions or set the plurality of positions of the left lung ribs according to actual needs. For example, the left and right lung ribs are symmetrical, each having twelve ribs, and the posterior ends of the ribs are connected to the thoracic vertebrae. The anterior ends of the upper five ribs are connected to the sternum. The anterior ends of the middle five ribs are fused into one rib and connected to the sternum. The anterior ends of the lower two ribs are free and combined to form the chest. Therefore, a person skilled in the art can configure the plurality of positions or set the plurality of positions of the left lung ribs, the plurality of positions or set the plurality of positions of the right lung ribs described below according to the actual positions of the ribs according to actual needs.

[0238] In embodiments of the present disclosure, the method for determining the plurality of positions of the left lung ribs comprises: performing rib detection on the lung registration costophrenic boundary and the pre-registration left lung costophrenic boundary respectively to obtain the plurality of positions of the left lung ribs.

[0239] In embodiments of the present disclosure, the method for performing rib detection on the lung registration costophrenic boundary and the pre-registration left lung costophrenic boundary respectively comprises: obtaining a left lung rib segmentation model corresponding to a preset convolutional neural network; performing rib detection on the lung registration costophrenic boundary and the pre-registration left lung costophrenic boundary based on the left lung rib segmentation model to obtain the plurality of positions of the left lung ribs.

[0240] In embodiments of the present disclosure, before performing rib detection on the lung registration costophrenic boundary and the pre-registration left lung costophrenic boundary based on the left lung rib segmentation model to obtain the plurality of positions of the left lung ribs, a left lung rib label corresponding to the left lung ribs is obtained, and the left lung rib segmentation model is trained based on the left lung rib label. The lung registration costophrenic boundary and the pre-registration left lung costophrenic boundary are detected based on the trained left lung rib segmentation model to obtain the plurality of positions of the left lung ribs.

[0241] In embodiments of the present disclosure, the method for determining the plurality of positions of the left lung ribs comprises: performing rib detection on the lung registration costophrenic boundary and the pre-registration left lung costophrenic boundary respectively to obtain the plurality of positions of the left lung ribs.

[0242] In embodiments of the present disclosure, the method for performing rib detection on the lung registration costophrenic boundary and the pre-registration left lung costophrenic boundary respectively comprises: obtaining a left lung rib segmentation model corresponding to a preset convolutional neural network; performing rib detection on the lung registration costophrenic boundary and the pre-registration left lung costophrenic boundary based on the left lung rib segmentation model to obtain the plurality of positions of the left lung ribs.

[0243] In embodiments of the present disclosure, the method for determining the second motion displacement according to the rib edge boundary in the right lung image corresponding to the plurality of DR lung images comprises: registering the rib edge boundary in the right lung image corresponding to the plurality of DR lung images to obtain a right lung registered rib edge boundary; and determining the second motion displacement according to the right lung registered rib edge boundary.

[0244] In embodiments of the present disclosure, the method for determining the second motion displacement according to the right lung registered rib edge boundary comprises: calculating a plurality of second distances corresponding to each boundary point in the right lung registered rib edge boundary and the right lung rib edge boundary before registration based on the right lung registered rib edge boundary; and configuring the plurality of second distances as the second motion displacement; or, the method for determining the second motion displacement according to the right lung registered rib edge boundary comprises: dividing the right lung registered rib edge boundary and the right lung rib edge boundary before registration into a second plurality of registered right lung rib edge boundaries and a second plurality of right lung rib edge boundaries before registration, respectively, based on a plurality of positions of the right lung rib or a plurality of set positions; and calculating a plurality of second distances corresponding to each boundary point in the second plurality of registered right lung rib edge boundaries and the second plurality of right lung rib edge boundaries before registration based on the right lung registered rib edge boundary, and configuring the plurality of second distances as the second motion displacement.

[0245] In embodiments of the present disclosure and other possible embodiments, the right lung registered rib edge boundary contains a one-to-one correspondence relationship (a registered boundary point pair) of each boundary point in the right lung registered rib edge boundary and the right lung rib edge boundary before registration. Therefore, the second distance of the position of each boundary point pair can be calculated, and all second distances of the positions of all boundary point pairs are the plurality of first distances.

[0246] In embodiments of the present disclosure and other possible embodiments, the plurality of first distances and the plurality of second distances can be configured as one or more of Euclidean distance, standardized Euclidean distance, Mahalanobis distance, Manhattan distance, Chebyshev distance, Minkowski distance, and Hamming distance.

[0247] In embodiments of the present disclosure, the method for determining the plurality of positions of the right lung rib comprises: performing rib detection on the right lung registered rib edge boundary and the right lung rib edge boundary before registration, respectively, to obtain the plurality of positions of the right lung rib.

[0248] In embodiments of the present disclosure, the method of performing rib detection on the right lung registered costophrenic boundary and the right lung costophrenic boundary before registration respectively comprises: obtaining a right lung rib segmentation model corresponding to a preset convolutional neural network; performing rib detection on the lung registered costophrenic boundary and the right lung costophrenic boundary before registration based on the right lung rib segmentation model to obtain a plurality of positions of the right lung ribs.

[0249] In embodiments of the present disclosure, before the rib detection on the lung registered costophrenic boundary and the right lung costophrenic boundary before registration based on the right lung rib segmentation model to obtain a plurality of positions of the right lung ribs, a right lung rib label corresponding to the right lung ribs is obtained, and the right lung rib segmentation model is trained based on the right lung rib label, and the rib detection on the lung registered costophrenic boundary and the right lung costophrenic boundary before registration is performed based on the trained right lung rib segmentation model to obtain a plurality of positions of the right lung ribs.

[0250] In embodiments of the present disclosure and other possible embodiments, the left lung rib segmentation model corresponding to the preset convolutional neural network and / or the right lung rib segmentation model corresponding to the preset convolutional neural network can be configured as a Unet convolutional neural network or a nnUnet convolutional neural network or a convolutional neural network improved based on the Unet convolutional neural network or a convolutional neural network improved based on the nnUnet convolutional neural network. For example, the convolutional neural network improved based on the Unet convolutional neural network can be configured as a ResUnet convolutional neural network with a residual structure.

[0251] In embodiments of the present disclosure and other possible embodiments, the Unet convolutional neural network or the nnUnet convolutional neural network or the convolutional neural network improved based on the Unet convolutional neural network or the convolutional neural network improved based on the nnUnet convolutional neural network at least includes a down-sampling contraction path, an up-sampling expansion path and a final classification layer.

[0252] In embodiments of the present disclosure and other possible embodiments, the registration method used by the present disclosure can use an existing registration algorithm or model, such as one or several of SIFT (Scale-invariant feature transform) registration algorithm or model, SURF (Speeded Up Robust Features) registration algorithm or model, ORB (Oriented FAST and Rotated BRIEF) registration algorithm or model, etc., or other registration algorithms or models based on convolutional neural networks. For example, the registration algorithm or model based on convolutional neural networks can be configured as a registration algorithm or model based on VGG network.

[0253] Step S103: determining whether the left lung and / or the right lung has lung adhesion based on the first motion displacement and / or the second motion displacement and a set motion displacement, respectively.

[0254] In embodiments of the present disclosure, the method of determining whether the left lung has lung adhesion based on the first motion displacement and a set motion displacement, respectively, includes: if the first motion displacement is less than or equal to the set motion displacement, determining that the left lung has lung adhesion; otherwise, the left lung does not have lung adhesion.

[0255] In embodiments of the present disclosure, the method of determining whether the right lung has lung adhesion based on the second motion displacement and a set motion displacement, respectively, includes: if the second motion displacement is less than or equal to the set motion displacement, determining that the right lung has lung adhesion; otherwise, the right lung does not have lung adhesion.

[0256] In embodiments of the present disclosure and other possible embodiments, the set motion displacement can be configured by those skilled in the art according to actual needs. For example, the set motion displacement can be configured as 1 mm, 1.2 mm, 1.5 mm, or other values.

[0257] In embodiments of the present disclosure and other possible embodiments, the first motion displacement includes a plurality of first distances; the second motion displacement includes a plurality of second distances; and determining whether the left lung and / or the right lung has lung adhesion based on the plurality of first distances of the first motion displacement and / or the plurality of second distances of the second motion displacement and a set motion displacement, respectively.

[0258] In embodiments of the present disclosure and other possible embodiments, the method of determining whether the left lung has lung adhesion based on the plurality of first distances of the first motion displacement and a set motion displacement, respectively, includes: if the plurality of first distances of the first motion displacement is less than or equal to the set motion displacement, determining that the left lung has lung adhesion; otherwise, the left lung does not have lung adhesion.

[0259] Similarly, in embodiments of the present disclosure and other possible embodiments, the method of determining whether the right lung has lung adhesion based on the plurality of second distances of the second motion displacement and a set motion displacement, respectively, includes: if the plurality of second distances of the second motion displacement is less than or equal to the set motion displacement, determining that the right lung has lung adhesion; otherwise, the right lung does not have lung adhesion.

[0260] In embodiments of the present disclosure and other possible embodiments, the left lung having lung adhesion or the right lung having lung adhesion means that the left lung or the right lung has adhesion with the costal margin boundary.

[0261] In embodiments of the present disclosure and other possible embodiments, if the left lung has lung adhesion, a first distance corresponding to the set motion displacement is extracted, and a coordinate corresponding to the first distance is configured as first position information of the left lung adhesion.

[0262] Similarly, in embodiments of the present disclosure and other possible embodiments, if the right lung has lung adhesion, a second distance corresponding to the set motion displacement is extracted, and a coordinate corresponding to the second distance is configured as second position information of the right lung adhesion.

[0263] In embodiments of the present disclosure and other possible embodiments, the method further comprises displaying the first position information of the left lung adhesion and / or the second position information of the right lung adhesion. The method of displaying the first position information of the left lung adhesion and / or the second position information of the right lung adhesion comprises displaying a plurality of DR lung images in a deep inhalation state and a deep exhalation state.

[0264] The first position information of the left lung adhesion and / or the second position information of the right lung adhesion is displayed on the basis of the displayed plurality of DR lung images in the deep inhalation state and the deep exhalation state.

[0265] In embodiments of the present disclosure and other possible embodiments, the method of displaying the first position information of the left lung adhesion and / or the second position information of the right lung adhesion comprises obtaining first set configuration information and / or second set configuration information. The first set configuration information and / or the second set configuration information respectively comprises a set line width and / or a set line color. The first set configuration information and / or the second set configuration information is respectively used to display the first position information of the left lung adhesion and / or the second position information of the right lung adhesion.

[0266] In embodiments of the present disclosure and other possible embodiments, a person skilled in the art can configure the set line width and / or the set line color according to actual needs. For example, the set line color can be configured as red or other colors.

[0267] In addition, the present disclosure also proposes an analysis method, which comprises or applies the lung adhesion detection method as described above to determine that the left lung has lung adhesion and / or the right lung has lung adhesion, and to evaluate whether functional disorders exist in diaphragmatic muscle movement corresponding to the plurality of DR lung images in the deep inhalation state and the deep exhalation state. If no functional disorders exist, an analysis result corresponding to the left lung having lung adhesion and / or the right lung having lung adhesion is given as a final result. Otherwise, an analysis result corresponding to the left lung not having lung adhesion and / or the right lung not having lung adhesion is given as a final result.

[0268] In embodiments of the present disclosure and other possible embodiments, the method for evaluating whether the diaphragm movement of a patient corresponding to a plurality of DR lung images in a deep inspiration state and a deep expiration state has a functional disorder, comprises: obtaining a set movement trajectory corresponding to a healthy person; detecting a diaphragm movement trajectory corresponding to a patient to be evaluated; and evaluating whether the diaphragm movement of the patient to be evaluated has a functional disorder based on the diaphragm movement trajectory corresponding to the patient to be evaluated and the set movement trajectory.

[0269] In embodiments of the present disclosure and other possible embodiments, the method for detecting the diaphragm movement trajectory corresponding to the patient to be evaluated, comprises: obtaining a plurality of two-dimensional DR left lung images and / or a plurality of two-dimensional DR right lung images at different times during the respiration of the patient to be evaluated; determining a plurality of left lung movement apices and / or a plurality of right lung movement apices respectively according to the plurality of two-dimensional DR left lung images and / or the plurality of two-dimensional DR right lung images; determining a plurality of left lung diaphragm movement point sequences and / or a plurality of right lung diaphragm movement point sequences respectively according to the plurality of two-dimensional DR left lung images and / or the plurality of two-dimensional DR right lung images; determining the movement of the left lung diaphragm based on the plurality of left lung movement apices and the plurality of left lung diaphragm movement point sequences, and / or determining the movement of the right lung diaphragm based on the plurality of right lung movement apices and the plurality of right lung diaphragm movement point sequences; determining the left lung diaphragm movement trajectory based on the plurality of left lung movement apices and the plurality of left lung diaphragm movement point sequences, and / or determining the right lung diaphragm movement trajectory based on the plurality of right lung movement apices and the plurality of right lung diaphragm movement point sequences; and wherein the set movement trajectory can be configured as the left lung diaphragm movement trajectory and / or the right lung diaphragm movement trajectory.

[0270] In embodiments of the present disclosure, before obtaining the plurality of two-dimensional DR left lung images and the plurality of two-dimensional DR right lung images at different times during the respiration, the plurality of two-dimensional DR lung images at different times during the respiration are segmented into left lung and right lung to obtain the plurality of two-dimensional DR left lung images and the plurality of two-dimensional DR right lung images.

[0271] In embodiments of the present disclosure, the method for segmenting the plurality of two-dimensional DR lung images at different times during the respiration into left lung and right lung to obtain the plurality of two-dimensional DR left lung images and the plurality of two-dimensional DR right lung images, comprises: detecting the rib boundary, the lung apex boundary, and the mediastinum and diaphragm edge of the left chest image and the right chest image of the plurality of two-dimensional DR lung images at different times during the respiration respectively to obtain the plurality of two-dimensional DR left lung images and the plurality of two-dimensional DR right lung images.

[0272] Or, in embodiments of the present disclosure, the method for left lung and right lung segmentation of the two-dimensional DR lung images at multiple time points in the breathing process comprises: obtaining a segmentation model of a preset convolutional neural network and a DR lung region label image used for training the segmentation model; wherein the method for determining the DR lung region label image used for training the segmentation model comprises: respectively performing rib border, lung apex boundary, and mediastinum and diaphragm edge detection on left chest images and right chest images of multiple DR lung region images to obtain DR lung region label images corresponding to the multiple DR lung region images; training the segmentation model using the DR lung region label images used for training the segmentation model; and based on the trained segmentation model, completing left lung and right lung segmentation of the two-dimensional DR lung images at multiple time points in the breathing process to obtain two-dimensional DR left lung images and two-dimensional DR right lung images at multiple time points.

[0273] In embodiments of the present disclosure and other possible embodiments, a to-be-processed DR image (a two-dimensional DR left lung image at multiple time points in the breathing process) is obtained, and it is determined whether the to-be-processed DR image is a lung image according to the to-be-processed DR image; wherein the lung image is configured as a two-dimensional DR lung image at multiple time points in the breathing process; if the to-be-processed DR image is the lung image, chest detection is performed on the to-be-processed DR image to remove information outside the chest in the to-be-processed DR image.

[0274] In embodiments of the present disclosure and other possible embodiments, the method for determining whether the to-be-processed DR image is a lung image comprises: calculating an average gray value corresponding to the to-be-processed DR image, and determining whether the to-be-processed DR image is a lung image according to the average gray value and a set gray value. Wherein, a person skilled in the art can configure the set gray value according to actual needs. For example, the set gray value can be configured as any one value or range between -1000HU and 0HU. Meanwhile, in embodiments of the present disclosure and other possible embodiments, the to-be-processed DR image can also be determined whether it is a lung image by manual judgment.

[0275] In embodiments of the present disclosure and other possible embodiments, the method for determining whether the DR image is a lung image according to the average gray value and a set gray value comprises: if the to-be-processed DR image is not subjected to the inverting processing, and if the average gray value is less than or equal to the set gray value, then it is determined that the to-be-processed DR image is a lung image; if the to-be-processed DR image is subjected to the inverting processing, and if the average gray value is greater than or equal to the set gray value, then it is determined that the to-be-processed DR image is a lung image. The principle is that if it is a lung image, the lung in the lung image is generally full of air or contains a certain amount of air, and the air corresponds to a small gray value (CT value), which is generally configured as -1000 HU; and the gray value (CT value) of water is generally configured as 0 HU, and the gray value (CT value) of bone is generally configured as 1000 HU or above; therefore, the above technical solution is used to determine whether the to-be-processed DR image is a lung image.

[0276] In embodiments of the present disclosure and other possible embodiments, if it is the lung image, then the to-be-processed DR image is subjected to the chest detection, and the information outside the chest in the to-be-processed DR image is removed. In embodiments of the present disclosure and other possible embodiments, the information outside the chest comprises: removing the unnecessary information such as arms, heads, and blank backgrounds.

[0277] In embodiments of the present disclosure, before the chest detection is performed on the to-be-processed DR image (a to-be-segmented lung image or a two-dimensional DR lung image at multiple moments in the breathing process or in the breath-holding state), the to-be-processed DR image is subjected to the filtering, and the to-be-segmented lung image after the filtering is subjected to the down-sampling to a set size.

[0278] In embodiments of the present disclosure, the logarithmic transformation of the to-be-processed DR image of the set size is subjected to the image enhancement, and an enhanced to-be-processed DR image is obtained.

[0279] In embodiments of the present disclosure, the ribcage image obtained by the chest detection of the to-be-processed DR image is configured as a to-be-segmented lung image; and based on the to-be-segmented lung image, the left lung and the right lung are segmented.

[0280] In embodiments of the present disclosure, the to-be-segmented lung image is segmented into a left chest image and a right chest image; and based on the left chest image and the right chest image respectively, the left lung and the right lung are segmented.

[0281] In the embodiments of the present disclosure, the lung image to be segmented is segmented into a left chest image and a right chest image; left lung and right lung segmentation is performed based on the left chest image and the right chest image respectively; or, a segmentation model of a preset convolutional neural network, a DR lung region label image used for training the segmentation model, and a plurality of DR lung images to be segmented (lung images to be segmented) at multiple moments in a breathing process or in a breath-holding state are obtained; wherein the method for determining the DR lung region label image used for training the segmentation model comprises: performing rib border, lung apex boundary, and mediastinum and diaphragm edge detection on the left chest image and the right chest image of the plurality of DR lung region images respectively to obtain the DR lung region label images corresponding to the plurality of DR lung region images; the DR lung region label image used for training the segmentation model is used to train the segmentation model; and based on the trained segmentation model, left lung and / or right lung segmentation of the plurality of DR lung images to be segmented is completed.

[0282] In the embodiments and other possible embodiments of the present disclosure, the segmentation model of the preset convolutional neural network is configured as a Unet convolutional neural network or a nnUnet convolutional neural network or a convolutional neural network improved based on the Unet convolutional neural network or a convolutional neural network improved based on the nnUnet convolutional neural network. For example, the convolutional neural network improved based on the Unet convolutional neural network can be configured as a ResUnet convolutional neural network with a residual structure.

[0283] In the embodiments and other possible embodiments of the present disclosure, the Unet convolutional neural network or the nnUnet convolutional neural network or the convolutional neural network improved based on the Unet convolutional neural network or the convolutional neural network improved based on the nnUnet convolutional neural network at least includes: a down-sampling contraction path, an up-sampling expansion path, and a final classification layer.

[0284] In the embodiments and other possible embodiments of the present disclosure, the plurality of DR lung region images are configured as a plurality of DR lung region images acquired in a deep inhalation state or in a breath-holding state.

[0285] In the embodiments and other possible embodiments of the present disclosure, before the segmentation model is trained by using the DR lung region label image used for training the segmentation model, data augmentation is performed on the DR lung region label image to obtain an augmented DR lung region label image; and the segmentation model is trained by using the augmented DR lung region label image.

[0286] In the embodiments and other possible embodiments of the present disclosure, a plurality of left lung motion apices and a plurality of right lung motion apices are determined according to the two-dimensional DR left lung images and the two-dimensional DR right lung images at multiple moments respectively.

[0287] In embodiments of the present disclosure, before the corresponding plurality of left lung motion vertices and the plurality of right lung motion vertices are determined according to the two-dimensional DR left lung images and the two-dimensional DR right lung images at different time points, respectively, and the plurality of left lung diaphragm motion point sequences and the plurality of right lung diaphragm motion point sequences are determined according to the two-dimensional DR left lung images and the two-dimensional DR right lung images at different time points, respectively, the method further comprises: extracting a plurality of left lung contour lines and a plurality of right lung contour lines corresponding to the two-dimensional DR left lung images and the two-dimensional DR right lung images at different time points, respectively; and determining the corresponding plurality of left lung motion vertices, the plurality of right lung motion vertices, the plurality of left lung diaphragm motion point sequences and the plurality of right lung diaphragm motion point sequences according to the plurality of left lung contour lines and the plurality of right lung contour lines.

[0288] In embodiments of the present disclosure and other possible embodiments, the method of extracting a plurality of left lung contour lines and a plurality of right lung contour lines corresponding to the two-dimensional DR left lung images and the two-dimensional DR right lung images at different time points, respectively, comprises: obtaining an edge detection algorithm or an edge detection model; and extracting a plurality of left lung contour lines and a plurality of right lung contour lines corresponding to the two-dimensional DR left lung images and the two-dimensional DR right lung images at different time points, respectively, by using the edge detection algorithm or the edge detection model.

[0289] In embodiments of the present disclosure and other possible embodiments, the edge detection algorithm or the edge detection model can be configured as one or more of edge detection algorithms or edge detection models based on Sobel operator, Prewitt operator, Roberts operator, Canny operator, Marr-Hildreth operator

[0290] In embodiments of the present disclosure, the method of determining the corresponding plurality of left lung motion vertices and the plurality of right lung motion vertices according to the plurality of left lung contour lines and the plurality of right lung contour lines comprises: detecting a plurality of left lung vertices and a plurality of right lung vertices corresponding to the plurality of left lung contour lines and the plurality of right lung contour lines, respectively; and configuring coordinates of the plurality of left lung vertices and the plurality of right lung vertices as the plurality of left lung motion vertices and the plurality of right lung motion vertices, respectively.

[0291] In embodiments of the present disclosure, the method of detecting a plurality of left lung top points and a plurality of right lung top points corresponding to the plurality of left lung contour lines and the plurality of right lung contour lines respectively comprises: determining a plurality of first maximum vertical coordinates and a plurality of second maximum vertical coordinates corresponding to the plurality of left lung contour lines and the plurality of right lung contour lines respectively; determining a plurality of first horizontal coordinates corresponding to the plurality of first maximum vertical coordinates according to the plurality of first maximum vertical coordinates and the plurality of left lung contour lines respectively; determining a plurality of second horizontal coordinates corresponding to the plurality of second maximum vertical coordinates according to the plurality of second maximum vertical coordinates and the plurality of right lung contour lines respectively; and configuring the plurality of first maximum vertical coordinates and the plurality of first horizontal coordinates corresponding thereto as the plurality of left lung top points respectively, and configuring the plurality of second maximum vertical coordinates and the plurality of second horizontal coordinates corresponding thereto as the plurality of right lung top points respectively.

[0292] In embodiments of the present disclosure and other possible embodiments, the method of determining a plurality of first maximum vertical coordinates and a plurality of second maximum vertical coordinates corresponding to the plurality of left lung contour lines and the plurality of right lung contour lines respectively comprises: establishing a coordinate system corresponding to the two-dimensional DR left lung images and the two-dimensional DR right lung images at multiple time points in the breathing process; and determining a plurality of first maximum vertical coordinates and a plurality of second maximum vertical coordinates corresponding to the plurality of left lung contour lines and the plurality of right lung contour lines respectively in the coordinate system.

[0293] In embodiments of the present disclosure, a plurality of left diaphragm motion point sequences and a plurality of right diaphragm motion point sequences are determined according to the two-dimensional DR left lung images and the two-dimensional DR right lung images at multiple time points respectively.

[0294] In embodiments of the present disclosure, the method of determining a plurality of left diaphragm motion point sequences and a plurality of right diaphragm motion point sequences corresponding to the plurality of left lung contour lines and the plurality of right lung contour lines respectively comprises: detecting a plurality of left lung lowest points and a plurality of right lung lowest points corresponding to the plurality of left lung contour lines and the plurality of right lung contour lines respectively, and configuring the plurality of left lung lowest points and the plurality of right lung lowest points as starting points of the plurality of left diaphragm motion point sequences and starting points of the plurality of right diaphragm motion point sequences respectively; detecting a plurality of left lung top points and a plurality of right lung top points corresponding to the plurality of left lung contour lines and the plurality of right lung contour lines respectively, and configuring coordinates corresponding to the plurality of left lung top points and the plurality of right lung top points as candidate end points of the plurality of left diaphragm motion point sequences and candidate end points of the plurality of right diaphragm motion point sequences respectively; determining the plurality of left diaphragm motion point sequences according to the starting points of the plurality of left diaphragm motion point sequences and the candidate end points of the plurality of left diaphragm motion point sequences respectively; and determining the plurality of right diaphragm motion point sequences according to the starting points of the plurality of right diaphragm motion point sequences and the candidate end points of the plurality of right diaphragm motion point sequences respectively.

[0295] Similarly, in embodiments of the present disclosure and other possible embodiments, in the coordinate system, the plurality of left diaphragm movement point sequences are determined according to the starting points of the plurality of left diaphragm movement point sequences and the candidate ending points of the plurality of left diaphragm movement point sequences, respectively.

[0296] In embodiments of the present disclosure, the method of determining the plurality of left diaphragm movement point sequences according to the starting points of the plurality of left diaphragm movement point sequences and the candidate ending points of the plurality of left diaphragm movement point sequences, respectively, includes: obtaining a first set step length; based on the first set step length, calculating a plurality of first slope values corresponding to the starting points of the plurality of left diaphragm movement point sequences to the candidate ending points of the plurality of left diaphragm movement point sequences, respectively; determining the ending points of the plurality of left diaphragm movement point sequences according to the plurality of first slope values; or, determining a left lung contour line between the starting points of the plurality of left diaphragm movement point sequences and the candidate ending points of the plurality of left diaphragm movement point sequences, respectively; and determining the ending points of the plurality of left diaphragm movement point sequences on the left lung contour line in an interactive manner.

[0297] In embodiments of the present disclosure and other possible embodiments, a person skilled in the art can configure the first set step length according to actual needs. For example, the first set step length can be configured as any value between 1-10 or other values.

[0298] For example, in a first starting point (x1, y1) of the starting points corresponding to the plurality of left diaphragm movement point sequences, the first set step length is configured as n, and the corresponding first slope value in the plurality of first slope values is (y 1+n -y1) / [x 1+n -x1), (y 1+2*n -y 1+n ) / [x 1+2*n -x 1+n ), ….

[0299] In embodiments of the present disclosure, the method of determining the ending points of the plurality of left diaphragm movement point sequences according to the plurality of first slope values includes: determining the candidate ending points corresponding to the change of the sign of the plurality of first slope values from negative to positive as the ending points of the plurality of left diaphragm movement point sequences.

[0300] In embodiments of the present disclosure and other possible embodiments, the method of determining the ending points of the plurality of left diaphragm movement point sequences on the left lung contour line in an interactive manner includes:

[0301] Color configuration is performed on the plurality of left lung contour lines between the starting points of the plurality of left diaphragm movement point sequences and the candidate ending points of the plurality of left diaphragm movement point sequences, respectively, and the plurality of left lung contour lines after color configuration are displayed, respectively;

[0302] Click the end points of the plurality of left diaphragm motion point sequences on the displayed plurality of left lung contour lines, respectively, to determine the end points of the plurality of left diaphragm motion point sequences, respectively.

[0303] In embodiments of the present disclosure, the method of determining the plurality of right diaphragm motion point sequences according to the start points of the plurality of right diaphragm motion point sequences and the candidate end points of the plurality of right diaphragm motion point sequences, respectively, comprises: obtaining a second set step length; calculating a plurality of second slope values corresponding to the start points of the plurality of right diaphragm motion point sequences to the candidate end points of the plurality of right diaphragm motion point sequences based on the second set step length; determining the end points of the plurality of right diaphragm motion point sequences according to the plurality of second slope values; or, determining the right lung contour lines between the start points of the plurality of right diaphragm motion point sequences and the candidate end points of the plurality of right diaphragm motion point sequences, respectively; and determining the end points of the plurality of right diaphragm motion point sequences on the right lung contour lines between, respectively, in an interactive manner.

[0304] In embodiments of the present disclosure and other possible embodiments, the second set step length can be configured by those skilled in the art according to actual needs. For example, the second set step length can be configured as any value between 1-10 or other values.

[0305] For example, in the first start point (x2, y2) among the start points corresponding to the plurality of right diaphragm motion point sequences, the second set step length is configured as m, and the corresponding second slope values in the plurality of second slope values are (y 2+m -y2) / [x 2+m -x2), (y 2+2*m -y 2+m ) / [x 2+2*m -x 2+m ), and so on.

[0306] In embodiments of the present disclosure, the method of determining the end points of the plurality of right diaphragm motion point sequences according to the plurality of second slope values comprises: determining the candidate end points corresponding to the change of the sign of the plurality of second slope values from positive to negative as the end points of the plurality of right diaphragm motion point sequences.

[0307] In embodiments of the present disclosure and other possible embodiments, the method of determining the end points of the plurality of right diaphragm motion point sequences on the right lung contour lines between, respectively, in an interactive manner comprises: color configuring a plurality of the right lung contour lines between, respectively, and displaying the plurality of right lung contour lines after color configuration, respectively; and clicking the end points of the plurality of right diaphragm motion point sequences on the displayed plurality of right lung contour lines, respectively, to determine the end points of the plurality of right diaphragm motion point sequences, respectively.

[0308] In embodiments of the present disclosure, the movement of the left lung diaphragm is determined based on the plurality of left lung movement peaks and the plurality of left lung diaphragm movement point sequences, and the movement of the right lung diaphragm is determined based on the plurality of right lung movement peaks and the plurality of right lung diaphragm movement point sequences.

[0309] In embodiments and other possible embodiments of the present disclosure, the method of determining the movement of the left lung diaphragm based on the plurality of left lung movement peaks and the plurality of left lung diaphragm movement point sequences includes: respectively configuring the plurality of left lung movement peaks as left lung movement reference points; based on the left lung movement reference points, respectively calculating a plurality of first distances between each movement point in the plurality of left lung diaphragm movement point sequences and the left lung movement reference points; or, respectively displaying the plurality of left lung movement peaks in first color lines, and respectively displaying the plurality of left lung diaphragm movement point sequences in second color lines.

[0310] In embodiments and other possible embodiments of the present disclosure, the method of determining the movement of the right lung diaphragm based on the plurality of right lung movement peaks and the plurality of right lung diaphragm movement point sequences includes: respectively configuring the plurality of right lung movement peaks as right lung movement reference points; based on the right lung movement reference points, respectively calculating a plurality of second distances between each movement point in the plurality of right lung diaphragm movement point sequences and the right lung movement reference points; or, respectively displaying the plurality of right lung movement peaks in first color lines, and respectively displaying the plurality of right lung diaphragm movement point sequences in second color lines.

[0311] In embodiments and other possible embodiments of the present disclosure, the method of displaying the plurality of left lung movement peaks and the plurality of right lung movement peaks in first color lines, and displaying the plurality of left lung diaphragm movement point sequences and the plurality of right lung diaphragm movement point sequences in second color lines includes: obtaining DR lung images corresponding to the DR left lung images and the DR right lung images at multiple times during the breathing process; displaying the DR lung images corresponding to the DR left lung images and the DR right lung images at multiple times during the breathing process; and displaying the plurality of left lung movement peaks and the plurality of right lung movement peaks in first color lines, and displaying the plurality of left lung diaphragm movement point sequences and the plurality of right lung diaphragm movement point sequences in second color lines on the DR lung images corresponding to the DR left lung images and the DR right lung images at multiple times during the breathing process.

[0312] In embodiments of the present disclosure and other possible embodiments, the method of displaying a plurality of left diaphragm movement point sequences and a plurality of right diaphragm movement point sequences in a second color line further includes: constructing a plurality of left diaphragm movement curves corresponding to the plurality of left diaphragm movement point sequences respectively, and configuring a movement point with the smallest slope of each left diaphragm movement curve in the plurality of left diaphragm movement curves as a left diaphragm movement point to be displayed at a plurality of moments in the breathing process; displaying the left diaphragm movement point to be displayed in the second color line; constructing a plurality of right diaphragm movement curves corresponding to the plurality of right diaphragm movement point sequences respectively, and configuring a movement point with the smallest slope of each right diaphragm movement curve in the plurality of right diaphragm movement curves as a right diaphragm movement point to be displayed at a plurality of moments in the breathing process; and displaying the right diaphragm movement point to be displayed in the second color line.

[0313] In embodiments of the present disclosure and other possible embodiments, the method of correcting the left diaphragm movement point to be displayed at a plurality of moments in the breathing process includes: obtaining a plurality of left lung areas in the breathing process; obtaining a plurality of first coordinate points corresponding to a first set slope value of each left diaphragm movement curve in the plurality of left diaphragm movement curves; sorting the plurality of first coordinate points corresponding to the first set slope value of each left diaphragm movement curve from small to large; and correcting the left diaphragm movement point to be displayed at the plurality of moments according to the plurality of left lung areas and the sorted plurality of first coordinate points.

[0314] In embodiments of the present disclosure and other possible embodiments, a person skilled in the art can configure the first set slope value according to actual needs. For example, the first set slope value can be configured as any value or other value between -0.2 and 0.

[0315] In embodiments of the present disclosure and other possible embodiments, the method of correcting the left diaphragm movement point to be displayed at a plurality of moments according to the plurality of left lung areas and the sorted plurality of first coordinate points includes: determining the sorted plurality of first coordinate points corresponding to the plurality of left lung areas from the largest area to the smallest area; wherein the ordinate values of the sorted plurality of first coordinate points corresponding to the plurality of left lung areas from the largest area to the smallest area are sequentially increased; if the ordinate of the left diaphragm movement point to be displayed at the plurality of moments corresponding to the plurality of left lung areas from the largest area to the smallest area is sequentially increased, the left diaphragm movement point to be displayed at the plurality of moments is not corrected; otherwise, the movement point not increasing in the left diaphragm movement point to be displayed at the plurality of moments is corrected.

[0316] In the embodiments of the present disclosure and other possible embodiments, the method for correcting the non-increasing motion points of the left lung diaphragm motion points to be displayed at the plurality of time points includes: extracting a first longitudinal coordinate corresponding to the non-increasing motion points of the plurality of left lung diaphragm motion points to be displayed at the plurality of time points and second and third longitudinal coordinates corresponding to two adjacent left lung diaphragm motion points to be displayed at adjacent time points on both sides of the non-increasing motion points; determining a first to-be-configured longitudinal coordinate greater than the third longitudinal coordinate and smaller than the second longitudinal coordinate (or greater than the second longitudinal coordinate and smaller than the third longitudinal coordinate) in the plurality of sorted first coordinate points; and configuring the first to-be-configured longitudinal coordinate and a transverse coordinate corresponding to the first to-be-configured longitudinal coordinate as the non-increasing motion points of the left lung diaphragm motion points to be displayed at the plurality of time points.

[0317] In the embodiments of the present disclosure and other possible embodiments, the method for correcting the right lung diaphragm motion points to be displayed at the plurality of time points during the breathing process includes: obtaining a plurality of right lung areas during the breathing process; obtaining a plurality of second coordinate points corresponding to a second set slope value of each right lung diaphragm motion curve in the plurality of right lung diaphragm motion curves; sorting the plurality of second coordinate points corresponding to the second set slope value of each right lung diaphragm motion curve in ascending order of longitudinal coordinates; and correcting the right lung diaphragm motion points to be displayed at the plurality of time points according to the plurality of right lung areas and the plurality of sorted second coordinate points.

[0318] In the embodiments of the present disclosure and other possible embodiments, the second set slope value can be configured by a person skilled in the art according to actual needs. For example, the second set slope value can be configured as any value or other value between 0 and 0.2.

[0319] In the embodiments of the present disclosure and other possible embodiments, the method for correcting the right lung diaphragm motion points to be displayed at the plurality of time points according to the plurality of right lung areas and the plurality of sorted second coordinate points includes: determining the plurality of sorted second coordinate points corresponding to the plurality of right lung areas from a maximum area to a minimum area; wherein the longitudinal coordinate values of the plurality of sorted second coordinate points corresponding to the plurality of right lung areas from the maximum area to the minimum area are sequentially increased; if the longitudinal coordinates of the plurality of right lung diaphragm motion points to be displayed at the plurality of time points are sequentially increased respectively with the plurality of sorted second coordinate points corresponding to the plurality of right lung areas from the maximum area to the minimum area, the right lung diaphragm motion points to be displayed at the plurality of time points are not corrected; otherwise, the non-increasing motion points of the right lung diaphragm motion points to be displayed at the plurality of time points are corrected.

[0320] In the embodiments and other possible embodiments of the present disclosure, the method for correcting the non-increasing motion points of the plurality of time instants of the to-be-displayed right lung diaphragm muscle motion points comprises: extracting a first longitudinal coordinate corresponding to the non-increasing motion points of the plurality of time instants of the to-be-displayed right lung diaphragm muscle motion points and second and third longitudinal coordinates corresponding to two adjacent to-be-displayed right lung diaphragm muscle motion points at adjacent time instants on both sides of the non-increasing motion points; determining a second to-be-configured longitudinal coordinate greater than the third longitudinal coordinate and smaller than the second longitudinal coordinate (or greater than the second longitudinal coordinate and smaller than the third longitudinal coordinate) in the sorted plurality of first coordinate points; and configuring the second to-be-configured longitudinal coordinate and a transverse coordinate corresponding thereto as the non-increasing motion points of the plurality of time instants of the to-be-displayed right lung diaphragm muscle motion points.

[0321] In the embodiments and other possible embodiments of the present disclosure, the method for displaying the plurality of left lung motion vertices and the plurality of right lung motion vertices in a first color line and displaying the plurality of left lung diaphragm muscle motion point sequences and the plurality of right lung diaphragm muscle motion point sequences in a second color line further comprises: obtaining first configuration parameters corresponding to the first color line; wherein the first configuration parameters comprise a first line length and a first line width; displaying the plurality of left lung motion vertices and the plurality of right lung motion vertices in the first color line configured with the first configuration parameters; obtaining first configuration parameters corresponding to the second color line; wherein the second configuration parameters comprise a second line length and a second line width; and displaying the plurality of left lung diaphragm muscle motion point sequences and the plurality of right lung diaphragm muscle motion point sequences in the second color line configured with the second configuration parameters. Wherein the first color is configured as red and the second color is configured as blue. In addition, the first color configuration and the second color configuration can be configured as lines of the same color. For example, red, blue or other colors.

[0322] In the embodiments and other possible embodiments of the present disclosure, the method for determining the diaphragm muscle motion trajectory corresponding to the to-be-evaluated patient and the set motion trajectory comprises: drawing a first diaphragm muscle motion trajectory corresponding to the plurality of left lung diaphragm muscle motion point sequences corresponding to the to-be-evaluated patient respectively; and / or drawing a second diaphragm muscle motion trajectory corresponding to the plurality of right lung diaphragm muscle motion point sequences corresponding to the to-be-evaluated patient respectively; and then determining the diaphragm muscle motion trajectory corresponding to the to-be-evaluated patient and the set motion trajectory.

[0323] In the embodiments and other possible embodiments of the present disclosure, before the set motion trajectory corresponding to the healthy person is obtained, the method for determining the set motion trajectory corresponding to the healthy person comprises: obtaining a plurality of diaphragm muscle motion trajectories corresponding to a preset number of healthy persons;

[0324] Fitting the plurality of diaphragm muscle motion trajectories to obtain the set motion trajectory corresponding to the healthy person.

[0325] Similarly, in the embodiments and other possible embodiments of the present disclosure, the set motion trajectory corresponding to the healthy person includes: drawing a first set motion trajectory corresponding to the left lung and / or a second set motion trajectory corresponding to the right lung of the healthy person respectively.

[0326] In the embodiments and other possible embodiments of the present disclosure, before the set motion trajectory corresponding to the healthy person is obtained, the method for determining the set motion trajectory corresponding to the healthy person includes: obtaining a plurality of left diaphragm motion trajectories and right diaphragm motion trajectories corresponding to the healthy person; fitting the plurality of left diaphragm motion trajectories to obtain a first set motion trajectory corresponding to the left lung of the healthy person; and fitting the plurality of right diaphragm motion trajectories to obtain a second set motion trajectory corresponding to the left lung of the healthy person.

[0327] In the embodiments and other possible embodiments of the present disclosure, the preset number can be configured by a person skilled in the art according to actual needs. At the same time, the calculation method of the similarity can be configured as one or more of cosine similarity (Cosine Similarity), adjusted cosine similarity (Adjusted Cosine Similarity), Pearson correlation coefficient (Pearson Correlation Coefficient), Jaccard similarity coefficient (Jaccard Coefficient), Tanimoto coefficient (generalized Jaccard similarity coefficient), log-likelihood similarity, mutual information / information gain, relative entropy / KL divergence, term frequency-inverse document frequency (TF-IDF) in information retrieval, etc.

[0328] In the embodiments and other possible embodiments of the present disclosure, based on the diaphragm motion trajectory corresponding to the patient to be evaluated and the set motion trajectory, the method for evaluating whether the diaphragm motion of the patient to be evaluated has a functional disorder includes: based on the first diaphragm motion trajectory and the first set motion trajectory, evaluating whether the left lung diaphragm motion of the patient to be evaluated has a functional disorder; and / or, based on the second diaphragm motion trajectory and the second set motion trajectory, evaluating whether the right lung diaphragm motion of the patient to be evaluated has a functional disorder.

[0329] In the embodiments of the present disclosure, the method for evaluating whether the diaphragm movement of the patient to be evaluated has functional disorders based on the diaphragm movement trajectory of the patient to be evaluated and the set movement trajectory comprises: calculating a similarity between the diaphragm movement trajectory and the set movement trajectory; if the similarity is less than a set similarity, the diaphragm movement of the patient to be evaluated has functional disorders; otherwise, the diaphragm movement of the patient to be evaluated does not have functional disorders.

[0330] In the embodiments of the present disclosure and other possible embodiments, the similarity of the set movement trajectory can be configured by those skilled in the art according to actual needs. For example, the similarity of the set movement trajectory can be configured as 0.8 or other numerical values.

[0331] In the embodiments of the present disclosure and other possible embodiments, the method for calculating the similarity between the diaphragm movement trajectory and the set movement trajectory, if the similarity is less than a set similarity, the diaphragm movement of the patient to be evaluated has functional disorders; otherwise, the diaphragm movement of the patient to be evaluated does not have functional disorders, comprises: calculating a first similarity between the first diaphragm movement trajectory and the first set movement trajectory, if the first similarity is less than a set similarity, the left lung diaphragm movement of the patient to be evaluated has functional disorders; otherwise, the left lung diaphragm movement of the patient to be evaluated does not have functional disorders; and / or, calculating a second similarity between the second diaphragm movement trajectory and the second set movement trajectory, if the second similarity is less than a set similarity, the right lung diaphragm movement of the patient to be evaluated has functional disorders; otherwise, the right lung diaphragm movement of the patient to be evaluated does not have functional disorders.

[0332] The execution subject of the lung adhesion detection method and the analysis method can be a lung adhesion detection device and an analysis device. For example, the lung adhesion detection method and the analysis method can be executed by a terminal device or a server or other processing device. The terminal device can be a user equipment (UE), a mobile device, a user terminal, a terminal, a cellular phone, a cordless phone, a personal digital assistant (PDA), a handheld device, a computing device, a vehicle-mounted device, a wearable device, and the like. In some possible implementation manners, the lung adhesion detection method and the analysis method can be implemented by a processor invoking computer-readable instructions stored in a memory.

[0333] Those skilled in the art can understand that, in the above lung adhesion detection method and analysis method of the specific embodiments, the writing order of each step does not mean a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of each step should be determined by its function and possible internal logic.

[0334] In addition, the disclosure embodiments also provide a lung adhesion detection device. The lung adhesion detection device comprises: an extraction unit configured to extract a rib edge boundary in a left lung image and / or a rib edge boundary in a right lung image corresponding to a plurality of DR lung images in a deep inhalation state and a deep exhalation state, respectively; a displacement determination unit configured to determine a first motion displacement according to the rib edge boundary in the left lung image corresponding to the plurality of DR lung images and / or determine a second motion displacement according to the rib edge boundary in the right lung image corresponding to the plurality of DR lung images; and a detection unit configured to determine whether the left lung and / or the right lung has lung adhesion based on the first motion displacement and / or the second motion displacement and a set motion displacement, respectively.

[0335] In addition, the disclosure embodiments also provide an analysis device. The analysis device comprises: a lung adhesion detection method as described above or a lung adhesion detection device as described above, and an evaluation unit configured to evaluate whether a diaphragm movement of a patient corresponding to the plurality of DR lung images in the deep inhalation state and the deep exhalation state has a functional disorder; and an analysis unit configured to give a final analysis result corresponding to that the left lung has lung adhesion and / or the right lung has lung adhesion if there is no functional disorder, or give a final analysis result corresponding to that the left lung has no lung adhesion and / or the right lung has no lung adhesion if there is a functional disorder.

[0336] In some embodiments, the device provided by the disclosure embodiments has functions or contains modules that can be used to perform the methods described in the above lung adhesion detection method and analysis method embodiments. The specific implementation can refer to the description of the above method embodiments, and for the sake of brevity, it will not be described here.

[0337] The disclosure embodiments also provide a computer-readable storage medium having stored thereon computer program instructions, which, when executed by a processor, implement the above lung adhesion detection method and / or analysis method. The computer-readable storage medium can be a non-volatile computer-readable storage medium.

[0338] The disclosure embodiments also provide an electronic device comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to implement the above lung adhesion detection method and / or analysis method. The electronic device can be provided as a terminal, a server or other forms of devices.

[0339] Figure 2 is a block diagram of an electronic device 800 according to an exemplary embodiment. For example, the electronic device 800 can be a terminal such as a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc.

[0340] Referring to Figure 2 The electronic device 800 can include one or more of the following components: a processing component 802, a memory 804, a power supply component 806, a multimedia component 808, an audio component 810, an input / output (I / O) interface 812, a sensor component 814, and a communication component 816.

[0341] The processing component 802 usually controls overall operations of the electronic device 800, such as operations associated with displaying, making phone calls, data communications, camera operations and recording operations. The processing component 802 can include one or more processors 820 to execute instructions to complete all or part of steps of the methods described above. In addition, the processing component 802 can include one or more modules to facilitate the interaction between the processing component 802 and other components. For example, the processing component 802 can include a multimedia module to facilitate the interaction between the multimedia component 808 and the processing component 802.

[0342] The memory 804 is configured to store various types of data to support operations of the electronic device 800. Examples of these data include instructions for any application or method operating on the electronic device 800, contact data, phonebook data, messages, pictures, videos, etc. The memory 804 can be implemented by any type of volatile or non-volatile storage devices or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0343] The power supply component 806 provides power to various components of the electronic device 800. The power supply component 806 can include a power supply management system, one or more power supplies, and other components associated with generating, managing and distributing power for the electronic device 800.

[0344] Multimedia component 808 includes a screen that provides an output interface between the electronic device 800 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may sense not only the boundaries of the touch or swipe action but also the duration and pressure associated with the touch or swipe operation. In some embodiments, multimedia component 808 includes a front-facing camera and / or a rear-facing camera. When the electronic device 800 is in an operating mode, such as a shooting mode or a video mode, the front-facing camera and / or the rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

[0345] Audio component 810 is configured to output and / or input audio signals. For example, audio component 810 includes a microphone (MIC) configured to receive external audio signals when electronic device 800 is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 804 or transmitted via communication component 816. In some embodiments, audio component 810 also includes a speaker for outputting audio signals.

[0346] I / O interface 812 provides an interface between processing component 802 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, power buttons, and lock buttons.

[0347] Sensor assembly 814 includes one or more sensors for providing state assessments of various aspects of electronic device 800. For example, sensor assembly 814 can detect the on / off state of electronic device 800, the relative positioning of components such as the display and keypad of electronic device 800, changes in position of electronic device 800 or a component of electronic device 800, the presence or absence of user contact with electronic device 800, orientation or acceleration / deceleration of electronic device 800, and temperature changes of electronic device 800. Sensor assembly 814 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 814 may also include an accelerometer, gyroscope, magnetometer, pressure sensor, or temperature sensor.

[0348] The communication component 816 is configured to facilitate wired or wireless communication between the electronic device 800 and other devices. The electronic device 800 can access a wireless network based on a communication standard, such as WiFi, 2G, or 3G, or a combination thereof. In an exemplary embodiment, the communication component 816 receives a broadcast signal or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 816 further includes a Near Field Communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on Radio Frequency Identification (RFID) techniques, infrared data association (IrDA) techniques, ultra-wideband (UWB) techniques, Bluetooth (BT) techniques, and other techniques.

[0349] In an exemplary embodiment, the electronic device 800 can be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, micro-controllers, microprocessors, or other electronic elements, for performing the above-described methods.

[0350] In an exemplary embodiment, a non-transitory computer-readable storage medium, such as the memory 804 including computer program instructions, is also provided, which can be executed by the processor 820 of the electronic device 800 to complete the above-described methods.

[0351] Figure 3 is a block diagram of an electronic device 1900 according to an exemplary embodiment. For example, the electronic device 1900 can be provided as a server. Referring to Figure 3 The electronic device 1900 includes a processing component 1922, which further includes one or more processors, and a memory resource represented by a memory 1932, for storing instructions, such as application programs, executable by the processing component 1922. The application programs stored in the memory 1932 can include one or more than one module each corresponding to a set of instructions. In addition, the processing component 1922 is configured to execute the instructions to perform the above-described methods.

[0352] The electronic device 1900 can also include a power supply component 1926 configured to perform power management of the electronic device 1900, a wired or wireless network interface 1950 configured to connect the electronic device 1900 to a network, and an input / output (I / O) interface 1958. The electronic device 1900 can operate based on an operating system stored in the memory 1932, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, or the like.

[0353] In example embodiments, a non-transitory computer-readable storage medium, e.g., memory 1932 including computer program instructions, is also provided that can be executed by processing component 1922 of electronic device 1900 to implement the above-described methods.

[0354] The present disclosure can be a system, a method, and / or a computer program product. The computer program product can include a computer readable storage medium (or media) having computer readable program instructions thereon for causing a processor to carry out aspects of the present disclosure.

[0355] The computer readable storage medium can be a tangible device that can retain and store instructions for use by an instruction execution device. The computer readable storage medium can be, for example, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. A non-exhaustive list of more specific examples of the computer readable storage medium includes the following: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanically encoded device such as punch-cards or punched tape, a

[0356] The computer readable program instructions described herein can be downloaded to respective computing / processing devices from a computer readable storage medium or to an external computer or external storage device via a network, for example, the Internet, a local area network, a wide area network and / or a wireless network. The network can comprise copper transmission cables, optical transmission fibers, wireless transmission, routers, firewalls, switches, gateway computers and / or edge servers. A network adapter card or network interface in each computing / processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium within the respective computing / processing device.

[0357] Computer readable program instructions for carrying out operations of the present disclosure can be assembler instructions, instruction-set-architecture (ISA) instructions, machine instructions, machine dependent instructions, microcode, firmware instructions, state-setting data, or either source code or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer readable program instructions can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider). In some embodiments, electronic circuitry including, for example, programmable logic circuitry, field-programmable gate array (FPGA), or programmable logic array (PLA) can execute the computer readable program instructions by utilizing state information of the computer readable program instructions to personalize the electronic circuitry, in order to perform aspects of the present disclosure.

[0358] The computer readable program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0359] These computer readable program instructions can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks. These computer readable program instructions can also be stored in a computer readable storage medium that can include random access memory (RAM), read only memory (ROM), electrically erasable programmable read only memory (EEPROM), flash memory or other data storage device. When the computer readable program instructions are loaded into the computer and other programmable data processing apparatus, a series of operational steps are implemented that provide processes such that the computer or other programmable apparatus provide processes for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0360] The computer readable program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus or other device to produce a computer implemented process such that the instructions which execute on the computer, other programmable data processing apparatus, or other device implement the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0361] The flow diagrams and the block diagrams in the drawings are presented to illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flow diagrams and the block diagrams can represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logic functions. In some alternative implementations, the functions noted in the blocks can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks can sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and / or flow diagrams, and combinations thereof, can be implemented by special purpose hardware-based systems that perform the specified functions or acts, or combinations of special purpose hardware and

[0362] Embodiments of the present disclosure have been described above, and the description is intended to be illustrative of the embodiments and not restrictive. Many modifications and variations of the described embodiments are possible and are within the scope of the disclosure. The selection of terms is intended to best describe the principles of the embodiments, practical application, or technical improvements over the technology found in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A lung adhesion detection method based on a DR image, characterized by, The method comprises the following steps: extracting one or both of the rib edge boundaries in the left lung image and the rib edge boundaries in the right lung image corresponding to the plurality of DR lung images in the deep inhalation state and the deep exhalation state respectively; determining a first motion displacement according to the rib edge boundaries in the left lung image corresponding to the plurality of DR lung images; determining whether the left lung has lung adhesion based on the first motion displacement and a set motion displacement; wherein the determination of the first motion displacement comprises: registering the rib edge boundaries in the left lung image corresponding to the plurality of DR lung images; calculating a plurality of first distances corresponding to each boundary point in the left lung rib edge boundaries before and after registration; and configuring the plurality of first distances as the first motion displacement; determining a second motion displacement according to the rib edge boundaries in the right lung image corresponding to the plurality of DR lung images; and determining whether the right lung has lung adhesion based on the second motion displacement and a set motion displacement; wherein the determination of the second motion displacement comprises: registering the rib edge boundaries in the right lung image corresponding to the plurality of DR lung images; calculating a plurality of second distances corresponding to each boundary point in the right lung rib edge boundaries before and after registration; and configuring the plurality of second distances as the second motion displacement.

2. The lung adhesion detection method according to claim 1, characterized by, The method comprises the following steps: extracting the rib edge boundaries in the left lung image corresponding to the plurality of DR lung images in the deep inhalation state and the deep exhalation state respectively, which comprises the following steps: constructing a directional derivative template of the left chest image corresponding to the plurality of DR lung images by using directional derivative, and setting a set weighted depth of the directional derivative template; performing template traversal of directional derivative on the left chest image by using the directional derivative template corresponding to the set weighted depth, superimposing the result of template traversal into the left chest image to obtain a left chest superimposed image; performing binaryzation processing on the left chest superimposed image to obtain a left rib edge binary image; obtaining a left rib edge angle image to be screened according to the left rib edge binary image and the left chest superimposed image; obtaining a screened left rib edge angle image based on the left rib edge angle image to be screened and a first set rib edge angle; performing connected domain selection on the left rib edge angle image to obtain a left lung image corresponding to a maximum connected domain.

3. The lung adhesion detection method according to claim 2, characterized by, The method comprises the following steps: performing morphological opening and closing operation and thinning processing on the left rib edge binary image to obtain a morphologically processed left rib edge binary image; performing AND operation on the gradient direction angle of each pixel in the morphologically processed left rib edge binary image and the left chest superimposed image to obtain the left rib edge angle image to be screened.

4. The lung adhesion detection method according to any one of claims 2 or 3, characterized in that, Before the step of constructing the directional derivative template of the left chest image corresponding to the plurality of DR lung images by using directional derivative, performing set scale Gaussian blur on the left chest image to obtain a corresponding left chest Gaussian blurred image; constructing a directional derivative template of the left chest Gaussian blurred image by using directional derivative; In the rib edge boundary extraction process of the left chest image, a direction derivative template corresponding to the set weighted depth is used to perform template traversal of the direction derivative of the left chest Gaussian blur image, the result of the template traversal is superimposed into the left chest image, and a left chest Gaussian blur superimposed image is obtained; The left chest Gaussian blur superimposed image is binarized to obtain a left rib edge binary image; and the left rib edge binary image and the left chest Gaussian blur superimposed image are used to obtain a left rib edge angle image to be screened.

5. The lung adhesion detection method according to any one of claims 1 to 3, characterized by, Comprise: Respectively extract rib edge boundaries in right lung images corresponding to a plurality of DR lung images under deep inspiration and deep expiration, comprising: A direction derivative template of a right chest image corresponding to the plurality of DR lung images is constructed by using a direction derivative, and a set weighted depth of the direction derivative template is set; A right chest superimposed image is obtained by using the direction derivative template corresponding to the set weighted depth to perform template traversal of the direction derivative of the right chest image, and superimposing the result of the template traversal into the right chest image; The right chest superimposed image is binarized to obtain a right rib edge binary image; A right rib edge angle image to be screened is obtained according to the right rib edge binary image and the right chest superimposed image; A screened right rib edge angle image is obtained based on the right rib edge angle image to be screened and a second set rib edge angle; The right rib edge angle image is subjected to connected domain selection to obtain a rib edge boundary in the right lung image corresponding to a maximum connected domain.

6. The lung adhesion detection method according to claim 4, characterized by, Comprise: Respectively extract rib edge boundaries in right lung images corresponding to a plurality of DR lung images under deep inspiration and deep expiration, comprising: A direction derivative template of a right chest image corresponding to the plurality of DR lung images is constructed by using a direction derivative, and a set weighted depth of the direction derivative template is set; A right chest superimposed image is obtained by using the direction derivative template corresponding to the set weighted depth to perform template traversal of the direction derivative of the right chest image, and superimposing the result of the template traversal into the right chest image; The right chest superimposed image is binarized to obtain a right rib edge binary image; A right rib edge angle image to be screened is obtained according to the right rib edge binary image and the right chest superimposed image; A screened right rib edge angle image is obtained based on the right rib edge angle image to be screened and a second set rib edge angle; The right rib edge angle image is subjected to connected domain selection to obtain a rib edge boundary in the right lung image corresponding to a maximum connected domain.

7. The lung adhesion detection method of claim 5, wherein, The right rib edge angle image to be screened is obtained according to the right rib edge binary image and the right chest superimposed image, comprising: The right rib edge binary image is subjected to morphological opening and closing operation and thinning processing to obtain a morphologically processed right rib edge binary image; The morphologically processed right rib edge binary image and the gradient direction angle of each pixel in the right chest superimposed image are subjected to AND operation to obtain the right rib edge angle image to be screened.

8. The lung adhesion detection method of claim 6, wherein, The right rib edge angle image to be screened is obtained according to the right rib edge binary image and the right chest superimposed image, comprising: Performing morphological open-close operation and thinning processing on the right rib edge binary image to obtain a morphologically processed right rib edge binary image; Performing AND operation on the gradient direction angle of each pixel in the morphologically processed right rib edge binary image and the right side chest superimposed image to obtain a right side rib edge angle image to be screened.

9. The lung adhesion detection method of claim 5, wherein, Before constructing the directional derivative template of the right side chest image corresponding to the plurality of DR lung images by using the directional derivative, performing Gaussian blur with a set scale on the right side chest image to obtain a corresponding right side chest Gaussian blur image; Constructing the directional derivative template of the right side chest Gaussian blur image by using the directional derivative; In the process of extracting the rib edge boundary of the right side chest image, performing template traversal of the directional derivative on the right side chest Gaussian blur image by using the directional derivative template corresponding to the set weighting depth, and superimposing the template traversal result into the right side chest image to obtain a right side chest Gaussian blur superimposed image; Performing binaryzation processing on the right side chest Gaussian blur superimposed image to obtain a right rib edge binary image; and obtaining a right side rib edge angle image to be screened according to the right rib edge binary image and the right side chest Gaussian blur superimposed image.

10. The lung adhesion detection method according to any one of claims 6-8, wherein, Before constructing the directional derivative template of the right side chest image corresponding to the plurality of DR lung images by using the directional derivative, performing Gaussian blur with a set scale on the right side chest image to obtain a corresponding right side chest Gaussian blur image; Constructing the directional derivative template of the right side chest Gaussian blur image by using the directional derivative; In the process of extracting the rib edge boundary of the right side chest image, performing template traversal of the directional derivative on the right side chest Gaussian blur image by using the directional derivative template corresponding to the set weighting depth, and superimposing the template traversal result into the right side chest image to obtain a right side chest Gaussian blur superimposed image; Performing binaryzation processing on the right side chest Gaussian blur superimposed image to obtain a right rib edge binary image; and obtaining a right side rib edge angle image to be screened according to the right rib edge binary image and the right side chest Gaussian blur superimposed image.

11. The lung adhesion detection method of claim 1, wherein, The method for screening the rib edge boundary of the right side chest image corresponding to the plurality of DR lung images comprises the following steps. Respectively extracting one or two kinds of rib edge boundaries in the left lung image and the right lung image corresponding to the plurality of DR lung images in the deep inspiration state and the deep expiration state, comprising: Respectively performing left lung and right lung segmentation on the plurality of DR lung images in the deep inspiration state and the deep expiration state to obtain one or two kinds of lung images in the left lung image and the right lung image corresponding to the plurality of DR lung images; 12. The lung adhesion detection method of claim 11, wherein, Respectively performing rib edge boundary detection on one or two kinds of lung images in the left lung image and the right lung image to obtain one or two kinds of rib edge boundaries in the left lung image and the right lung image corresponding to the plurality of DR lung images. The method for screening the rib edge boundary of the right side chest image corresponding to the plurality of DR lung images comprises the following steps. Respectively performing rib edge boundary detection on one or two kinds of lung images in the left lung image and the right lung image to obtain one or two kinds of rib edge boundaries in the left lung image and the right lung image corresponding to the plurality of DR lung images, comprising: extracting a plurality of left lung contour lines and a plurality of right lung contour lines corresponding to one or two lung images of the left lung image and the right lung image corresponding to the plurality of DR lung images respectively; determining one or two rib edge boundaries in the left lung image and the right lung image corresponding to the plurality of DR lung images according to one or two lung contour lines of the plurality of left lung contour lines and the plurality of right lung contour lines respectively.

13. The lung adhesion detection method of claim 12, wherein, The method for determining one or two rib edge boundaries in the left lung image and the right lung image corresponding to the plurality of DR lung images according to one or two lung contour lines of the plurality of left lung contour lines and the plurality of right lung contour lines respectively comprises: determining a plurality of left lung motion peaks and a plurality of left lung motion lowest points corresponding to the plurality of left lung contour lines according to the plurality of left lung contour lines; and determining the rib edge boundary in the left lung image corresponding to the plurality of DR lung images based on the plurality of left lung motion peaks and the plurality of left lung motion lowest points; determining a plurality of right lung motion peaks and a plurality of right lung motion lowest points corresponding to the plurality of right lung contour lines according to the plurality of right lung contour lines; and determining the rib edge boundary in the right lung image corresponding to the plurality of DR lung images based on the plurality of right lung motion peaks and the plurality of right lung motion lowest points.

14. The lung adhesion detection method of claim 13, wherein, The method for determining the rib edge boundary in the left lung image corresponding to the plurality of DR lung images based on the plurality of left lung motion peaks and the plurality of left lung motion lowest points comprises: connecting the plurality of left lung motion peaks and the plurality of left lung motion lowest points corresponding thereto to obtain a plurality of first division lines; configuring the rib edge boundary on the right side of the plurality of first division lines in the left lung image as the rib edge boundary in the left lung image corresponding to the plurality of DR lung images.

15. The lung adhesion detection method according to any one of claims 13 or 14, wherein, The method for determining the rib edge boundary in the right lung image corresponding to the plurality of DR lung images based on the plurality of right lung motion peaks and the plurality of right lung motion lowest points comprises: connecting the plurality of right lung motion peaks and the plurality of right lung motion lowest points corresponding thereto to obtain a plurality of second division lines; configuring the rib edge boundary on the left side of the plurality of second division lines in the right lung image as the rib edge boundary in the right lung image corresponding to the plurality of DR lung images.

16. The lung adhesion detection method according to any one of claims 11-14, wherein, The method for performing left lung and right lung segmentation on the plurality of DR lung images in the deep inhalation state and the deep exhalation state respectively to obtain the left lung image and the right lung image corresponding to the plurality of DR lung images comprises: performing rib edge boundary, lung apex boundary and mediastinum and diaphragm edge detection on left chest images and right chest images of the plurality of DR lung images in the deep inhalation state and the deep exhalation state respectively to obtain the left lung image and the right lung image corresponding to the plurality of DR lung images; or obtaining a segmentation model of a preset convolutional neural network and a DR lung region label image used for training the segmentation model; training the segmentation model by using the DR lung region label image used for training the segmentation model; and completing left lung and right lung segmentation of the plurality of DR lung images in the deep inhalation state and the deep exhalation state based on the trained segmentation model to obtain the left lung image and the right lung image corresponding to the plurality of DR lung images.

17. The lung adhesion detection method of claim 15, wherein, The left lung and right lung segmentation of the plurality of DR lung images in the deep inspiration state and the deep expiration state respectively, to obtain the left lung images and the right lung images corresponding to the plurality of DR lung images, comprising: The rib edge boundary, lung apex boundary and mediastinum and diaphragm edge detection of the left chest image and the right chest image of the plurality of DR lung images in the deep inspiration state and the deep expiration state respectively, to obtain the left lung images and the right lung images corresponding to the plurality of DR lung images; or, Obtain the segmentation model of the preset convolutional neural network and the DR lung area label image for training the segmentation model; train the segmentation model by using the DR lung area label image for training the segmentation model; based on the trained segmentation model, complete the left lung and right lung segmentation of the plurality of DR lung images in the deep inspiration state and the deep expiration state, to obtain the left lung images and the right lung images corresponding to the plurality of DR lung images.

18. The lung adhesion detection method according to any one of claims 1-3, 6-9, 11-14, wherein, The plurality of first distances corresponding to each boundary point in the left lung rib edge boundary before and after registration are calculated; The plurality of first distances are configured as first motion displacements, comprising: Based on a plurality of positions of the left lung ribs or setting a plurality of positions, the left lung rib edge boundary before and after registration of the left lung is divided respectively to obtain a first plurality of segmented left lung rib edges before registration and a first plurality of segmented left lung rib edges after registration; The plurality of first distances corresponding to each boundary point in the first plurality of segmented left lung rib edges before registration and the first plurality of segmented left lung rib edges after registration are calculated respectively.

19. The lung adhesion detection method of claim 4, wherein, The plurality of first distances corresponding to each boundary point in the left lung rib edge boundary before and after registration are calculated; The plurality of first distances are configured as first motion displacements, comprising: Based on a plurality of positions of the left lung ribs or setting a plurality of positions, the left lung rib edge boundary before and after registration of the left lung is divided respectively to obtain a first plurality of segmented left lung rib edges before registration and a first plurality of segmented left lung rib edges after registration; The plurality of first distances corresponding to each boundary point in the first plurality of segmented left lung rib edges before registration and the first plurality of segmented left lung rib edges after registration are calculated respectively.

20. The lung adhesion detection method of claim 5, wherein, The plurality of first distances corresponding to each boundary point in the left lung rib edge boundary before and after registration are calculated; The plurality of first distances are configured as first motion displacements, comprising: Based on a plurality of positions of the left lung ribs or setting a plurality of positions, the left lung rib edge boundary before and after registration of the left lung is divided respectively to obtain a first plurality of segmented left lung rib edges before registration and a first plurality of segmented left lung rib edges after registration; The plurality of first distances corresponding to each boundary point in the first plurality of segmented left lung rib edges before registration and the first plurality of segmented left lung rib edges after registration are calculated respectively.

21. The lung adhesion detection method of claim 10, wherein, The plurality of first distances corresponding to each boundary point in the left lung rib edge boundary before and after registration are calculated; The plurality of first distances are configured as first motion displacements, comprising: Based on a plurality of positions of the left lung ribs or setting a plurality of positions, the left lung rib edge boundary before and after registration of the left lung is divided respectively to obtain a first plurality of segmented left lung rib edges before registration and a first plurality of segmented left lung rib edges after registration; The plurality of first distances corresponding to each boundary point in the first plurality of segmented left lung rib edges before registration and the first plurality of segmented left lung rib edges after registration are calculated respectively.

22. The lung adhesion detection method of claim 15, wherein, The calculation involves multiple first distances corresponding to each boundary point in the left lung costal margin boundary before and after registration; configuring these multiple first distances as first motion displacements, including: Based on multiple locations of the left lung ribs or multiple locations set, the left lung rib margin boundary is divided before and after registration to obtain the first multi-segment left lung rib margin edge before registration and the first multi-segment left lung rib margin edge after registration. Calculate the multiple first distances corresponding to each boundary point in the left lung costal margin before and after the first multi-segment registration.

23. The lung adhesion detection method of claim 16, wherein, The calculation involves multiple first distances corresponding to each boundary point in the left lung costal margin boundary before and after registration. Configuring the plurality of first distances as first motion displacements includes: Based on multiple locations of the left lung ribs or multiple locations set, the left lung rib margin boundary is divided before and after registration to obtain the first multi-segment left lung rib margin edge before registration and the first multi-segment left lung rib margin edge after registration. Calculate the multiple first distances corresponding to each boundary point in the left lung costal margin before and after the first multi-segment registration.

24. The lung adhesion detection method of claim 18, wherein, Determine multiple locations of the left lung rib, including: The left lung ribs were detected by examining both the registered rib margin boundary and the unregistered left lung rib margin boundary, resulting in multiple locations of the left lung ribs.

25. The lung adhesion detection method of any one of claims 19-23, wherein, Determine multiple locations of the left lung rib, including: The left lung ribs were detected by examining both the registered rib margin boundary and the unregistered left lung rib margin boundary, resulting in multiple locations of the left lung ribs.

26. The lung adhesion detection method of claim 24, wherein, The method involves rib detection on both the registered left lung rib margin boundary and the pre-registration left lung rib margin boundary to obtain multiple locations of the left lung ribs, including: Obtain the left lung rib segmentation model corresponding to the preset convolutional neural network; Based on the left lung rib segmentation model, rib detection is performed on the registered rib margin boundary and the left lung rib margin boundary before registration to obtain multiple positions of the left lung ribs.

27. The lung adhesion detection method of claim 25, wherein, The method involves rib detection on both the registered left lung rib margin boundary and the pre-registration left lung rib margin boundary to obtain multiple locations of the left lung ribs, including: Obtain the left lung rib segmentation model corresponding to the preset convolutional neural network; Based on the left lung rib segmentation model, rib detection is performed on the registered rib margin boundary and the left lung rib margin boundary before registration to obtain multiple positions of the left lung ribs.

28. The lung adhesion detection method according to any one of claims 26 or 27, wherein, Before performing rib detection on the lung registration rib margin boundary and the pre-registration left lung rib margin boundary based on the left lung rib segmentation model to obtain multiple positions of the left lung ribs, the following steps are included: Obtain the left lung rib label corresponding to the left lung rib; The left lung rib segmentation model is trained based on the left lung rib tags; Based on the trained left lung rib segmentation model, rib detection is performed on the registered rib margin boundary and the unregistered left lung rib margin boundary to obtain multiple locations of the left lung ribs.

29. The lung adhesion detection method of any one of claims 1-3, 6-9, 11-14, 17, 19-24, 26, 27, wherein, The calculation of multiple second distances corresponding to each boundary point in the right lung costal margin boundary before and after registration includes: Based on multiple locations of the right lung ribs or by setting multiple locations, the right lung rib margin boundary is divided both before and after registration to obtain the second multi-segment right lung rib margin edge before registration and the second multi-segment right lung rib margin edge after registration. Second distances corresponding to each boundary point in the second multi-segment pre-registration right lung costal margin edge and the second multi-segment post-registration right lung costal margin edge are respectively calculated.

30. The lung adhesion detection method of claim 4, wherein, The calculating of the second distances corresponding to each boundary point in the pre-registration and post-registration right lung costal margin boundaries comprises: Based on the positions of the right lung ribs or setting positions, the right lung post-registration and pre-registration right lung costal margin boundaries are respectively divided to obtain the second multi-segment pre-registration right lung costal margin edge and the second multi-segment post-registration right lung costal margin edge; Second distances corresponding to each boundary point in the second multi-segment pre-registration right lung costal margin edge and the second multi-segment post-registration right lung costal margin edge are respectively calculated.

31. The lung adhesion detection method of claim 5, wherein, The calculating of the second distances corresponding to each boundary point in the pre-registration and post-registration right lung costal margin boundaries comprises: Based on the positions of the right lung ribs or setting positions, the right lung post-registration and pre-registration right lung costal margin boundaries are respectively divided to obtain the second multi-segment pre-registration right lung costal margin edge and the second multi-segment post-registration right lung costal margin edge; Second distances corresponding to each boundary point in the second multi-segment pre-registration right lung costal margin edge and the second multi-segment post-registration right lung costal margin edge are respectively calculated.

32. The lung adhesion detection method of claim 10, wherein, The calculating of the second distances corresponding to each boundary point in the pre-registration and post-registration right lung costal margin boundaries comprises: Based on the positions of the right lung ribs or setting positions, the right lung post-registration and pre-registration right lung costal margin boundaries are respectively divided to obtain the second multi-segment pre-registration right lung costal margin edge and the second multi-segment post-registration right lung costal margin edge; Second distances corresponding to each boundary point in the second multi-segment pre-registration right lung costal margin edge and the second multi-segment post-registration right lung costal margin edge are respectively calculated.

33. The lung adhesion detection method of claim 15, wherein, The calculating of the second distances corresponding to each boundary point in the pre-registration and post-registration right lung costal margin boundaries comprises: Based on the positions of the right lung ribs or setting positions, the right lung post-registration and pre-registration right lung costal margin boundaries are respectively divided to obtain the second multi-segment pre-registration right lung costal margin edge and the second multi-segment post-registration right lung costal margin edge; Second distances corresponding to each boundary point in the second multi-segment pre-registration right lung costal margin edge and the second multi-segment post-registration right lung costal margin edge are respectively calculated.

34. The lung adhesion detection method of claim 16, wherein, The calculating of the second distances corresponding to each boundary point in the pre-registration and post-registration right lung costal margin boundaries comprises: Based on the positions of the right lung ribs or setting positions, the right lung post-registration and pre-registration right lung costal margin boundaries are respectively divided to obtain the second multi-segment pre-registration right lung costal margin edge and the second multi-segment post-registration right lung costal margin edge; Second distances corresponding to each boundary point in the second multi-segment pre-registration right lung costal margin edge and the second multi-segment post-registration right lung costal margin edge are respectively calculated.

35. The lung adhesion detection method of claim 18, wherein, The calculating of the second distances corresponding to each boundary point in the pre-registration and post-registration right lung costal margin boundaries comprises: Based on the positions of the right lung ribs or setting positions, the right lung post-registration and pre-registration right lung costal margin boundaries are respectively divided to obtain the second multi-segment pre-registration right lung costal margin edge and the second multi-segment post-registration right lung costal margin edge; Second distances corresponding to each boundary point in the second multi-segment pre-registration right lung costal margin edge and the second multi-segment post-registration right lung costal margin edge are respectively calculated. The calculating of the second distances corresponding to each boundary point in the pre-registration and post-registration right lung costal margin boundaries comprises: Based on the positions of the right lung ribs or setting positions, the right lung post-registration and pre-registration right lung costal margin boundaries are respectively divided to obtain the second multi-segment pre-registration right lung costal margin edge and the second multi-segment post-registration right lung costal margin edge; Second distances corresponding to each boundary point in the second multi-segment pre-registration right lung costal margin edge and the second multi-segment post-registration right lung costal margin edge are respectively calculated.

36. The lung adhesion detection method of claim 25, wherein, The second distances corresponding to each boundary point in the right lung rib edge boundary before and after registration are calculated, and the second distances include: A plurality of positions of the right lung rib are determined or set, and the right lung rib edge boundary after registration and the right lung rib edge boundary before registration are divided respectively to obtain a second multi-segment right lung rib edge before registration and a second multi-segment right lung rib edge after registration. The second distances corresponding to each boundary point in the second multi-segment right lung rib edge before registration and the second multi-segment right lung rib edge after registration are calculated respectively.

37. The lung adhesion detection method of claim 28, wherein, The second distances corresponding to each boundary point in the right lung rib edge boundary before and after registration are calculated, and the second distances include: A plurality of positions of the right lung rib are determined or set, and the right lung rib edge boundary after registration and the right lung rib edge boundary before registration are divided respectively to obtain a second multi-segment right lung rib edge before registration and a second multi-segment right lung rib edge after registration. The second distances corresponding to each boundary point in the second multi-segment right lung rib edge before registration and the second multi-segment right lung rib edge after registration are calculated respectively.

38. The lung adhesion detection method of claim 29, wherein, The plurality of positions of the right lung rib are determined, and the plurality of positions include: Rib detection is performed on the right lung rib edge boundary after registration and the right lung rib edge boundary before registration respectively to obtain the plurality of positions of the right lung rib.

39. The lung adhesion detection method of any one of claims 30-37, wherein, The plurality of positions of the right lung rib are determined, and the plurality of positions include: Rib detection is performed on the right lung rib edge boundary after registration and the right lung rib edge boundary before registration respectively to obtain the plurality of positions of the right lung rib.

40. The lung adhesion detection method of claim 38, wherein, The rib detection on the right lung rib edge boundary after registration and the right lung rib edge boundary before registration respectively includes: A right lung rib segmentation model corresponding to a preset convolutional neural network is acquired. Rib detection is performed on the right lung rib edge boundary after registration and the right lung rib edge boundary before registration based on the right lung rib segmentation model to obtain the plurality of positions of the right lung rib.

41. The lung adhesion detection method of claim 39, wherein, The rib detection on the right lung rib edge boundary after registration and the right lung rib edge boundary before registration respectively includes: A right lung rib segmentation model corresponding to a preset convolutional neural network is acquired. Rib detection is performed on the right lung rib edge boundary after registration and the right lung rib edge boundary before registration based on the right lung rib segmentation model to obtain the plurality of positions of the right lung rib.

42. The lung adhesion detection method of any one of claims 40 or 41, wherein, Before the rib detection on the right lung rib edge boundary after registration and the right lung rib edge boundary before registration based on the right lung rib segmentation model to obtain the plurality of positions of the right lung rib, the method further includes: A right lung rib label corresponding to the right lung rib is acquired. The right lung rib segmentation model is trained based on the right lung rib label. Rib detection is performed on the right lung rib edge boundary after registration and the right lung rib edge boundary before registration based on the trained right lung rib segmentation model to obtain the plurality of positions of the right lung rib.

43. The lung adhesion detection method of any one of claims 1-3, 6-9, 11-14, 17, 19-24, 26, 27, 30-38, 40, 41, wherein, The determination of whether the left lung has lung adhesion based on the first motion displacement and a set motion displacement respectively includes: If the first motion displacement is less than or equal to the set motion displacement, it is determined that the left lung has lung adhesion; otherwise, the left lung does not have lung adhesion.

44. The lung adhesion detection method of claim 4, wherein, The determination of whether the left lung has lung adhesion based on the first motion displacement and a set motion displacement respectively includes: If the first motion displacement is less than or equal to the set motion displacement, it is determined that the left lung has lung adhesion; otherwise, the left lung does not have lung adhesion.

45. The lung adhesion detection method of claim 5, wherein, The determining whether the left lung has lung adhesion based on the first motion displacement and a set motion displacement respectively comprises: If the first motion displacement is less than or equal to the set motion displacement, it is determined that the left lung has lung adhesion; otherwise, the left lung does not have lung adhesion.

46. The lung adhesion detection method of claim 10, wherein, The determining whether the left lung has lung adhesion based on the first motion displacement and a set motion displacement respectively comprises: If the first motion displacement is less than or equal to the set motion displacement, it is determined that the left lung has lung adhesion; otherwise, the left lung does not have lung adhesion.

47. The lung adhesion detection method of claim 15, wherein, The determining whether the left lung has lung adhesion based on the first motion displacement and a set motion displacement respectively comprises: If the first motion displacement is less than or equal to the set motion displacement, it is determined that the left lung has lung adhesion; otherwise, the left lung does not have lung adhesion.

48. The lung adhesion detection method of claim 16, wherein, The determining whether the left lung has lung adhesion based on the first motion displacement and a set motion displacement respectively comprises: If the first motion displacement is less than or equal to the set motion displacement, it is determined that the left lung has lung adhesion; otherwise, the left lung does not have lung adhesion.

49. The lung adhesion detection method of claim 18, wherein, The determining whether the left lung has lung adhesion based on the first motion displacement and a set motion displacement respectively comprises: If the first motion displacement is less than or equal to the set motion displacement, it is determined that the left lung has lung adhesion; otherwise, the left lung does not have lung adhesion.

50. The lung adhesion detection method of claim 25, wherein, The determining whether the left lung has lung adhesion based on the first motion displacement and a set motion displacement respectively comprises: If the first motion displacement is less than or equal to the set motion displacement, it is determined that the left lung has lung adhesion; otherwise, the left lung does not have lung adhesion.

51. The lung adhesion detection method of claim 28, wherein, The determining whether the left lung has lung adhesion based on the first motion displacement and a set motion displacement respectively comprises: If the first motion displacement is less than or equal to the set motion displacement, it is determined that the left lung has lung adhesion; otherwise, the left lung does not have lung adhesion.

52. The lung adhesion detection method of claim 29, wherein, The determining whether the left lung has lung adhesion based on the first motion displacement and a set motion displacement respectively comprises: If the first motion displacement is less than or equal to the set motion displacement, it is determined that the left lung has lung adhesion; otherwise, the left lung does not have lung adhesion.

53. The lung adhesion detection method of claim 39, wherein, The determining whether the left lung has lung adhesion based on the first motion displacement and a set motion displacement respectively comprises: If the first motion displacement is less than or equal to the set motion displacement, it is determined that the left lung has lung adhesion; otherwise, the left lung does not have lung adhesion.

54. The lung adhesion detection method of claim 42, wherein, The determining whether the left lung has lung adhesion based on the first motion displacement and a set motion displacement respectively comprises: If the first motion displacement is less than or equal to the set motion displacement, it is determined that the left lung has lung adhesion; otherwise, the left lung does not have lung adhesion.

55. The lung adhesion detection method of any one of claims 1-3, 6-9, 11-14, 17, 19-24, 26, 27, 30-38, 40, 41, 44-54, wherein, The determining whether the left lung has lung adhesion based on the first motion displacement and a set motion displacement respectively comprises: If the first motion displacement is less than or equal to the set motion displacement, it is determined that the left lung has lung adhesion; otherwise, the left lung does not have lung adhesion.

56. The lung adhesion detection method of claim 4, wherein, The determining whether the left lung has lung adhesion based on the first motion displacement and a set motion displacement respectively comprises: If the first motion displacement is less than or equal to the set motion displacement, it is determined that the left lung has lung adhesion; otherwise, the left lung does not have lung adhesion. The determining whether the right lung has lung adhesion based on the second motion displacement and a set motion displacement respectively comprises: If the second motion displacement is less than or equal to the set motion displacement, it is determined that the right lung has lung adhesion; otherwise, the right lung does not have lung adhesion. The determining whether the right lung has lung adhesion based on the second motion displacement and a set motion displacement respectively comprises: If the second motion displacement is less than or equal to the set motion displacement, it is determined that the right lung has lung adhesion; otherwise, the right lung does not have lung adhesion. If the second motion displacement is less than or equal to the set motion displacement, it is determined that the right lung has lung adhesion; otherwise, the right lung does not have lung adhesion.

57. The lung adhesion detection method of claim 5, wherein, The determination of whether the right lung has lung adhesion based on the second motion displacement and the set motion displacement respectively includes: If the second motion displacement is less than or equal to the set motion displacement, it is determined that the right lung has lung adhesion; otherwise, the right lung does not have lung adhesion.

58. The lung adhesion detection method of claim 10, wherein, The determination of whether the right lung has lung adhesion based on the second motion displacement and the set motion displacement respectively includes: If the second motion displacement is less than or equal to the set motion displacement, it is determined that the right lung has lung adhesion; otherwise, the right lung does not have lung adhesion.

59. The lung adhesion detection method of claim 15, wherein, The determination of whether the right lung has lung adhesion based on the second motion displacement and the set motion displacement respectively includes: If the second motion displacement is less than or equal to the set motion displacement, it is determined that the right lung has lung adhesion; otherwise, the right lung does not have lung adhesion.

60. The lung adhesion detection method of claim 16, wherein, The determination of whether the right lung has lung adhesion based on the second motion displacement and the set motion displacement respectively includes: If the second motion displacement is less than or equal to the set motion displacement, it is determined that the right lung has lung adhesion; otherwise, the right lung does not have lung adhesion.

61. The lung adhesion detection method of claim 18, wherein, The determination of whether the right lung has lung adhesion based on the second motion displacement and the set motion displacement respectively includes: If the second motion displacement is less than or equal to the set motion displacement, it is determined that the right lung has lung adhesion; otherwise, the right lung does not have lung adhesion.

62. The lung adhesion detection method of claim 25, wherein, The determination of whether the right lung has lung adhesion based on the second motion displacement and the set motion displacement respectively includes: If the second motion displacement is less than or equal to the set motion displacement, it is determined that the right lung has lung adhesion; otherwise, the right lung does not have lung adhesion.

63. The lung adhesion detection method of claim 28, wherein, The determination of whether the right lung has lung adhesion based on the second motion displacement and the set motion displacement respectively includes: If the second motion displacement is less than or equal to the set motion displacement, it is determined that the right lung has lung adhesion; otherwise, the right lung does not have lung adhesion.

64. The lung adhesion detection method of claim 29, wherein, The determination of whether the right lung has lung adhesion based on the second motion displacement and the set motion displacement respectively includes: If the second motion displacement is less than or equal to the set motion displacement, it is determined that the right lung has lung adhesion; otherwise, the right lung does not have lung adhesion.

65. The lung adhesion detection method of claim 39, wherein, The determination of whether the right lung has lung adhesion based on the second motion displacement and the set motion displacement respectively includes: If the second motion displacement is less than or equal to the set motion displacement, it is determined that the right lung has lung adhesion; otherwise, the right lung does not have lung adhesion.

66. The lung adhesion detection method of claim 42, wherein, The determination of whether the right lung has lung adhesion based on the second motion displacement and the set motion displacement respectively includes: If the second motion displacement is less than or equal to the set motion displacement, it is determined that the right lung has lung adhesion; otherwise, the right lung does not have lung adhesion.

67. The lung adhesion detection method of claim 43, wherein, The determination of whether the right lung has lung adhesion based on the second motion displacement and the set motion displacement respectively includes: If the second motion displacement is less than or equal to the set motion displacement, it is determined that the right lung has lung adhesion; otherwise, the right lung does not have lung adhesion. The determination of whether the right lung has lung adhesion based on the second motion displacement and the set motion displacement respectively includes: If the second motion displacement is less than or equal to the set motion displacement, it is determined that the right lung has lung adhesion; otherwise, the right lung does not have lung adhesion. The determination of whether the right lung has lung adhesion based on the second motion displacement and the set motion displacement respectively includes: If the second motion displacement is less than or equal to the set motion displacement, it is determined that the right lung has lung adhesion; otherwise, the right lung does not have lung adhesion.

68. An assay method characterized by, The lung adhesion detection method according to any one of claims 1-67 is included or applied to determine that the left lung has lung adhesion and / or the right lung has lung adhesion; and, evaluate whether there is a functional disorder in the diaphragmatic movement of the patient corresponding to the multiple DR lung images in the deep inspiration state and the deep expiration state; if there is no functional disorder, give the final analysis result corresponding to that the left lung has lung adhesion and / or the right lung has lung adhesion; otherwise, give the final analysis result corresponding to that the left lung does not have lung adhesion and / or the right lung does not have lung adhesion.

69. A lung adhesion detection device, comprising: comprise: an extraction unit configured to extract one or both of the rib edge boundaries in the left lung images and the rib edge boundaries in the right lung images corresponding to the multiple DR lung images in the deep inspiration state and the deep expiration state, respectively; a displacement determination unit configured to determine a first movement displacement based on the rib edge boundaries in the left lung images corresponding to the multiple DR lung images; and determine a second movement displacement based on the rib edge boundaries in the right lung images corresponding to the multiple DR lung images; a detection unit configured to determine whether the left lung has lung adhesion based on the first movement displacement and a set movement displacement, wherein the determination of the first movement displacement comprises: registering the rib edge boundaries in the left lung images corresponding to the multiple DR lung images; calculating a plurality of first distances corresponding to each boundary point in the registered and unregistered left lung rib edge boundaries; and configuring the plurality of first distances as the first movement displacement; and determine whether the right lung has lung adhesion based on the second movement displacement and the set movement displacement, wherein the determination of the second movement displacement comprises: registering the rib edge boundaries in the right lung images corresponding to the multiple DR lung images; calculating a plurality of second distances corresponding to each boundary point in the registered and unregistered right lung rib edge boundaries; and configuring the plurality of second distances as the second movement displacement.

70. An assay device, characterized by The lung adhesion detection device according to claim 69 is included or applied to determine that the left lung has lung adhesion and / or the right lung has lung adhesion; and, evaluate whether there is a functional disorder in the diaphragmatic movement of the patient corresponding to the multiple DR lung images in the deep inspiration state and the deep expiration state; if there is no functional disorder, give the final analysis result corresponding to that the left lung has lung adhesion and / or the right lung has lung adhesion; otherwise, give the final analysis result corresponding to that the left lung does not have lung adhesion and / or the right lung does not have lung adhesion. comprise:

71. An electronic device, comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to invoke the instructions stored in the memory to perform the lung adhesion detection method according to any one of claims 1-67. comprise:

72. An electronic device, comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to invoke the instructions stored in the memory to perform the analysis method according to claim 68. The computer program instructions, when executed by the processor, implement the lung adhesion detection method according to any one of claims 1-67.

73. A computer-readable storage medium having stored thereon computer program instructions, wherein, The computer program instructions, when executed by the processor, implement the analysis method according to claim 68. 74.A computer readable storage medium having stored thereon computer program instructions, wherein, The computer program instructions, when executed by the processor, implement the lung adhesion detection method according to any one of claims 1-67. The computer program instructions, when executed by the processor, implement the analysis method according to claim 68.

Images