CN-121767509-B - CT image reconstruction and bronchoscope image fusion system based on trachea walking

CN121767509BCN 121767509 BCN121767509 BCN 121767509BCN-121767509-B

Abstract

The invention relates to the technical field of image processing, and discloses a CT image reconstruction and bronchoscope image fusion system based on trachea walking, which firstly extracts three-dimensional structures of a trachea and a bronchus from a chest CT original image, and generates an intracavity visual angle CT reconstruction image consistent with the airway walking based on a trachea center line; the system further acquires a real-time bronchoscope video stream, spatial registration of an intracavity visual angle CT image and the real-time bronchoscope video is achieved through three-dimensional coordinate mapping, the registered CT image and the bronchoscope video are fused and rendered in a gray substrate and color coverage mode to obtain a fused image with mucous membrane details and structural background, and the system can also plan a bronchoscope insertion path based on the intracavity CT image and support interactive association display among an original CT image, a three-dimensional tracheal tree model and the intracavity visual angle CT image. The invention can form a planning path before bronchoscopy, provide real-time structure guidance in the process, and form a multi-mode fusion image after completion.

Inventors

MA LIJIE
WANG HUANYU
ZHONG YUJIE
WEI WENXIANG

Assignees

中国人民解放军西部战区总医院

Dates

Publication Date: 20260512
Application Date: 20260305

Claims (10)

1. CT image rebuilds and bronchoscope image fusion system based on trachea walking, its characterized in that includes: The tracheal tree extraction module is used for extracting three-dimensional structure data of the trachea and bronchi at all levels based on input chest CT original image data, including but not limited to airway radius and branch angles and branch topological relations, and establishing a three-dimensional tracheal tree model containing a tracheal center line; The CT reconstruction module is connected with the tracheal tree extraction module and is used for adaptively adjusting the fault plane angle along the central line and always perpendicular to the central line by taking the tracheal central line of the three-dimensional tracheal tree model as a reference to generate an intracavity visual angle CT image consistent with the running of the trachea, and the CT reconstruction module is concretely based on a central line parameter curve Calculating tangent vector at any point The cutting vector is used for defining the running direction of the air pipe and can be obtained by deriving a central line: ; Wherein, the Is a parameter of arc length A unit tangent vector at the position representing the local running direction of the trachea at the position; , , respectively at the central line , , Derivative of direction; At each of the Creating a local orthogonal coordinate system Wherein In order to cut the vector quantity, 、 Is two mutually orthogonal and are both with Orthogonal normal vectors; in the above-mentioned local orthogonal coordinate system, the reconstruction plane of the intracavity view CT image can be defined as passing through the centerline point The normal vector is Plane of plane (c) of its coordinates And world coordinates The mapping relationship between them can be expressed as: ; Wherein, the Reconstructing coordinates of points on a plane in a three-dimensional space; , Are respectively along the local coordinates in the plane And Offset of direction; Three-dimensional coordinates of the central line point of the position by the method Uniformly sample on plane and will Mapping back to original CT volume data The three-linear interpolation is performed, so that a sequential intracavity visual angle tomographic image along the running direction of the trachea can be generated; The bronchoscope video stream data input module is used for acquiring a bronchoscope real-time video stream; The gray substrate uploading module is used for uploading the reconstructed CT image data to the bronchoscopy system; The image matching module is connected with the CT reconstruction module and the bronchoscope video stream data input module and is used for realizing the matching and alignment of the spatial positions between the intracavity visual angle CT image and the real-time video stream based on a three-dimensional space coordinate mapping method; the fusion rendering module is connected with the image matching module and is used for carrying out fusion rendering on the registered intracavity visual angle CT image and the real-time video stream, wherein the gray structure of the intracavity visual angle CT image is a basal layer, and the color image of the real-time video stream is a covering layer to be overlapped and presented; And the fusion video output module is connected with the fusion rendering module and is used for outputting the video content after fusion rendering and generating an inspection report.
2. The CT image reconstruction and bronchoscope image fusion system based on tracheal walking of claim 1, further comprising a navigation module connected with the CT reconstruction module for automatically generating an optimal path schematic or dynamic navigation video for bronchoscope insertion based on the intra-cavity view CT image analysis airway geometry and walking direction.
3. The system for reconstructing CT images and merging bronchoscopic images based on tracheal walking of claim 1, further comprising a CT redisplay module connected with the tracheal tree extraction module and the CT reconstruction module for performing position correlation between the three-dimensional tracheal tree model and the intracavity view CT image and the original chest CT image, so as to interactively retrieve the intracavity view CT image at the corresponding position on the original chest CT image.
4. The tracheal walking-based CT image reconstruction and bronchoscope image fusion system according to claim 1, wherein the image matching module realizes spatial registration of intra-cavity view CT images and real-time video streams by a three-dimensional coordinate mapping method based on an airway centerline coordinate system.
5. The tracheal walking-based CT image reconstruction and bronchoscope image fusion system of claim 1, wherein the tracheal tree extraction module extracts three-dimensional tracheal structures from chest CT raw image data using region growing methods, morphological operations, or a combination of both.
6. The tracheal walking-based CT image reconstruction and bronchoscope image fusion system of claim 1, wherein the CT reconstruction module generates the intra-cavity view CT image using a multi-planar reconstruction technique, a curved surface reconstruction technique, or a centerline-based curved surface unfolding reconstruction technique.
7. The CT image reconstruction and bronchoscope image fusion method based on trachea walking is characterized by comprising the following steps of: s1, extracting a three-dimensional tracheal tree model based on chest CT original image data and acquiring a tracheal central line; S2, taking the tracheal central line as a reconstruction reference, adaptively adjusting the reconstruction plane angle along the running direction of the central line, and generating a cavity view CT image, wherein the method specifically comprises the following steps: Based on a centreline parameter curve Calculating tangent vector at any point The cutting vector is used for defining the running direction of the air pipe and can be obtained by deriving a central line: ; Wherein, the Is a parameter of arc length A unit tangent vector at the position representing the local running direction of the trachea at the position; , , respectively at the central line , , Derivative of direction; At each of the Creating a local orthogonal coordinate system Wherein In order to cut the vector quantity, 、 Is two mutually orthogonal and are both with Orthogonal normal vectors; in the above-mentioned local orthogonal coordinate system, the reconstruction plane of the intracavity view CT image can be defined as passing through the centerline point The normal vector is Plane of plane (c) of its coordinates And world coordinates The mapping relationship between them can be expressed as: ; Wherein, the Reconstructing coordinates of points on a plane in a three-dimensional space; , Are respectively along the local coordinates in the plane And Offset of direction; Three-dimensional coordinates of the central line point of the position by the method Uniformly sample on plane and will Mapping back to original CT volume data The three-linear interpolation is performed, so that a sequential intracavity visual angle tomographic image along the running direction of the trachea can be generated; S3, collecting real-time video streams of the bronchoscope; s4, based on a three-dimensional coordinate mapping method, anatomical position matching is carried out on the intracavity visual angle CT image and the real-time video stream; s5, fusion rendering is carried out on the matched intracavity view angle CT image and the real-time video stream, wherein the former is used as a gray base layer, and the latter is used as a color covering layer; s6, outputting the video after fusion rendering as an inspection report.
8. The method of claim 7, further comprising planning a bronchoscope insertion path and generating navigation information based on the intra-cavity view CT image generated in step S2.
9. The method of claim 7, further comprising coordinate correlating the three-dimensional airway tree model and the intra-cavity view CT image with the original chest CT image such that the original CT image interactively invokes the intra-cavity view CT image at the corresponding location.
10. A computer-readable storage medium, characterized in that a computer program is stored thereon, which program, when being executed by a processor, implements the method steps of any of claims 7 to 9.

Description

CT image reconstruction and bronchoscope image fusion system based on trachea walking Technical Field The invention relates to the technical field of image processing, in particular to a CT image reconstruction and bronchoscope image fusion system based on trachea walking. Background In the existing clinical respiratory intervention diagnosis and treatment, bronchoscopy is an important means for acquiring airway lesion information and implementing biopsy and treatment. However, because the tracheal tree has a complex structure, numerous branches and obvious individual differences, doctors often need to determine the turning direction and the insertion depth depending on experience in the bronchoscopy process, especially in the area with narrow peripheral bronchial paths and high bending degree, the clinical operation difficulty is high, and the focus arrival rate is limited. The traditional bronchoscope vision can only display color images of the mucous membrane surface, and cannot present anatomical structures of tissues outside and around the airway wall, so that global space reference is lost during deep navigation, and the risk of mistakenly entering a wrong branch or failing to proceed is increased. With the development of medical imaging technology, chest CT has become an important fundamental image for lung lesion assessment. Technologies such as three-dimensional reconstruction based on CT data, virtual bronchoscopy and airway tree analysis are gradually used for bronchoscopy navigation assistance. However, most of traditional CT reconstructions adopt fixed planes such as axial, coronal, sagittal, etc., or realize the integral display of the airway structure through curved surface reconstruction, and the image view angle has obvious difference from the intracavity view angle of the actual bronchoscope, so that it is difficult to provide anatomical reference consistent with the endoscope view field. Although the virtual bronchoscope can simulate the intracavity visual angle, the image of the virtual bronchoscope is derived from CT gray data, and has obvious difference with the color mucous membrane texture of the real bronchoscope, so that doctors often need to frequently switch between the two systems when using the virtual bronchoscope, and real-time integrated reference is difficult to realize. On the other hand, the position of the real-time video of the bronchoscope and the CT image are relatively difficult to correspond. The bronchoscope imaging process has the characteristics of dynamic property, rapid visual angle change, unstable illumination condition and the like, the CT image is a static gray image, and the CT image and the static gray image have essential differences in space coordinates, imaging principles and information expression forms, so that direct alignment and fusion between the CT image and the static gray image are difficult to realize in the traditional method. Some existing navigation systems mainly rely on pre-calculated tracheal tree paths or three-dimensional models to prompt a doctor for directions, but cannot provide real-time image display, and cannot dynamically present CT tissue backgrounds matched with the current field of view in the actual endoscopic operation process, so that limitations still exist in deep navigation of complex bronchial trees. In clinical practice, there is another problem that when a doctor reads an original chest CT image, a doctor cannot intuitively determine a specific anatomical position corresponding to a specific slice position in a view angle in a bronchoscope cavity, and the "view angle fault" causes a lack of correlation between the CT image and an endoscope operation, which affects accuracy of diagnosis and navigation. The lack of a system that can establish an interactive correspondence between the original CT image, the three-dimensional model of the airway tree, and the endoscopic view angle also constitutes a deficiency of the prior art. Therefore, there is a strong need for CT image reconstruction and bronchoscopic image fusion systems based on tracheal walking to solve the above problems. Disclosure of Invention The invention aims to solve the technical problems in the background technology and provides a CT image reconstruction and bronchoscope image fusion system based on tracheal walking. The above object of the present invention is achieved by: CT image rebuilds and bronchoscope image fusion system based on trachea walking includes: The tracheal tree extraction module is used for extracting three-dimensional structure data of the trachea and bronchi at all levels according to input chest CT original image data, including but not limited to airway radius, branch angles and branch topological relations, and establishing a three-dimensional tracheal tree model containing a tracheal center line; The CT reconstruction module is connected with the tracheal tree extraction module, takes the tracheal center line of the three-dimens